Part2_Google_Maps_Analysis_public_version.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "d04111ef",
   "metadata": {},
   "source": [
    "## Google Maps | Location History Analysis\n",
    "\n",
    "<img src=\"https://images.unsplash.com/photo-1548345680-f5475ea5df84?ixlib=rb-4.0.3&ixid=MnwxMjA3fDB8MHxwaG90by1wYWdlfHx8fGVufDB8fHx8&auto=format&fit=crop&w=2073&q=80\" alt=\"Google Maps image\" width=\"500\" style=\"margin-left:0\"/>    \n",
    "\n",
    "Credits: Photo by <a href=\"https://unsplash.com/@hjkp?utm_source=unsplash&utm_medium=referral&utm_content=creditCopyText\">henry perks</a> on <a href=\"https://unsplash.com/s/photos/google-maps?utm_source=unsplash&utm_medium=referral&utm_content=creditCopyText\">Unsplash</a>\n",
    "  \n",
    "\n",
    "*Project summary:*\n",
    "In my [Samsung Health project](https://github.com/MargotMarchais/Samsung_Health_analysis), I analyzed several years of my pedometer data (number of steps and calories burned). Yet, these data do not include all my workout activities. For instance, swimming and dance classes are not present in the data because I only own a smartphone (i.e. not a smartwatch). Thus, the analysis of my Samsung Health data underestimates my total activity workouts by focusing only on activities that can be easily recorded with a smartphone like running and hiking.\n",
    "\n",
    "To bridge the gap, I used a second data source: Google Maps Location History data. With those data, I could identify the places where I used to go swimming and dancing, and the times I went there. To do so, I imported dozens of JSON files and loaded them into 2 Python dataframes. The aim of the project is to track down the number of times I went to the swimming pool / dance studio and the average duration of my workouts.\n",
    "\n",
    "## TOC:\n",
    "* [1. Import the relevant packages](#1)   \n",
    "* [2. Convert dozens of JSON files into 2 Python lists](#2)\n",
    "* [3. Transform the lists into Pandas dataframes](#3)\n",
    "* [4. Clean data](#4)\n",
    "* [5. Enrich data](#5)\n",
    " * [5.1. Dates and durations](#5.1)\n",
    " * [5.2. Add manually names to addresses with no Google Maps name](#5.2)\n",
    " * [5.3. Extract City, Zipcode and Country information](#5.3)\n",
    "* [6. Export the raw data to csv](#6)\n",
    "* [7. Data visualization and insights](#7)\n",
    " * [7.1 Number of times I did swimming and dancing workouts](#7.1)\n",
    " * [7.2 Workout durations](#7.2)\n",
    " * [7.3 Places visited during my workouts](#7.3)\n",
    "* [8. Conclusion](#8)\n",
    "* [8.1 Key learnings](#8.1)\n",
    "* [8.2 Next steps](#8.2)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e51d69d6",
   "metadata": {},
   "source": [
    "<div style=\"background-color: RGB(0,150,250);\" >\n",
    "<h2 style=\"margin: auto; padding: 20px; color:#fff; \">1. Import the relevant packages and modules</h2> <a class=\"anchor\" id=\"1\"></a>\n",
    "</div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "d7ed8771",
   "metadata": {},
   "outputs": [],
   "source": [
    "from zipfile import ZipFile\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import json\n",
    "import datetime\n",
    "import seaborn as sns\n",
    "import string"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7e6f585f",
   "metadata": {},
   "source": [
    "<div style=\"background-color: RGB(0,150,250);\" >\n",
    "<h2 style=\"margin: auto; padding: 20px; color:#fff; \">2. Convert dozens of JSON files into 2 Python lists</h2> <a class=\"anchor\" id=\"2\"></a>\n",
    "</div>\n",
    "\n",
    "This step is without any doubt the most difficult part in the project. Indeed, when downloaded, Google Maps data are:\n",
    "- located in different subfolders (one per year)\n",
    "- stored across multiple files (I received 72 files in total)\n",
    "- and written in deep-nested JSON format (which was a new format to me)\n",
    "\n",
    "To import the data, I tackled the places visited and the activity segments separately (2 chunks of code, see below).\n",
    "\n",
    "Note: For this step, I am very grateful to the software engineer [Maksym Kozlenko](https://betterprogramming.pub/loading-location-history-places-from-google-timeline-into-pandas-and-csv-c26cb0ac5e89) who shared his tricks to import all JSON data into a single file. I reused the code and enriched it to also take into consideration the activity segments (see below)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "b9208755",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Objective: Import all the JSON data related to place visited (respectively activities) into a single file\n",
    "\n",
    "# Provide the path to my Google Location History takeout file\n",
    "history_data_file = \"C:/Users/margo/Documents/Documents/Formation/Github/Google_Maps_Location_History/takeout-20230806T160809Z-001.zip\"\n",
    "\n",
    "# Store all places visited into the array called \"place_visits\", and all activities into the array called \"activity_segments\".\n",
    "# For now, we set both arrays as empty.\n",
    "place_visits = []\n",
    "activity_segments = []\n",
    "\n",
    "# Google Maps data are obtained through a zip file (cf. folder \"takeout-20230806T160809Z-001\" mentioned in the history_data_file variable). \n",
    "    # In this zip file, there are many folders and files that we want to open and read their information.\n",
    "with ZipFile(history_data_file) as myzip:\n",
    "    for file in myzip.filelist[:]:\n",
    "        filename = file.filename\n",
    "        \n",
    "        # we process all files in the \"Semantic Location History\" directory (this is where all the JSON files are located)\n",
    "        if \"Semantic Location History\" in filename:\n",
    "            history_json = json.load(myzip.open(filename))  # We load every JSON file found in the zip folder\n",
    "\n",
    "            # The JSON files always start with the \"timelineObjects\" JSON object\n",
    "            for timeline_object in history_json[\"timelineObjects\"]:\n",
    "               \n",
    "            # Step 1 : Extract information about places visited\n",
    "                # placeVisit is one of the most important Google Maps objects. It is located in timelineObjects.\n",
    "                if \"placeVisit\" in timeline_object:\n",
    "                    place_visit_json = timeline_object[\"placeVisit\"]\n",
    "                    \n",
    "                    # skip places without geographical coordinates\n",
    "                    if not \"location\" in place_visit_json or not \"latitudeE7\" in place_visit_json[\"location\"]:\n",
    "                        continue\n",
    "                    \n",
    "                    # We specify each column we want to extract for our future places_visit dataframe. \n",
    "                        # Note: We identify them thanks to their position in the JSON file.\n",
    "                    place_visit = {\n",
    "                        \"placeId\": place_visit_json[\"location\"][\"placeId\"],\n",
    "                        \"locationConfidence\": place_visit_json[\"location\"][\"locationConfidence\"],\n",
    "                        \"startTimestamp\": place_visit_json[\"duration\"][\"startTimestamp\"],\n",
    "                        \"endTimestamp\": place_visit_json[\"duration\"][\"endTimestamp\"],\n",
    "                        \"placeVisitImportance\": place_visit_json[\"placeVisitImportance\"],\n",
    "                        \"placeVisitType\": place_visit_json[\"placeVisitType\"],\n",
    "                        \"latitudeE7\": place_visit_json[\"location\"][\"latitudeE7\"],\n",
    "                        \"longitudeE7\": place_visit_json[\"location\"][\"longitudeE7\"],\n",
    "                    }\n",
    "                    \n",
    "                    # Below, we treat differently special columns like centerLatE7, centerLngE7, name and address\n",
    "                        # Indeed, these columns do not always contain information so we offer the possibility that values are Null (\"None\")\n",
    "                        \n",
    "                    # Optional columns in the timelineObjects object\n",
    "                    for optional_field in [\"centerLatE7\", \"centerLngE7\"]:\n",
    "                        if optional_field in place_visit_json:\n",
    "                            place_visit[optional_field] = place_visit_json[optional_field] # path to find the information if it exists\n",
    "                        else:\n",
    "                            place_visit[optional_field] = None # in case it does not exist, we set it to None\n",
    "                    \n",
    "                    # Optional columns in the timelineObjects / location object\n",
    "                    for optional_field in [\"name\", \"address\"]:\n",
    "                        if optional_field in place_visit_json[\"location\"]:\n",
    "                            place_visit[optional_field] = place_visit_json[\"location\"][optional_field] # path to find the information if it exists\n",
    "                        else:\n",
    "                            place_visit[optional_field] = None # in case it does not exist, we set it to None\n",
    "                        \n",
    "                    # Final step: Concatenate the information from all files into the array place_visits\n",
    "                    place_visits.append(place_visit)\n",
    "                    \n",
    "                    \n",
    "                # Step 2 : Extract information from activities performed\n",
    "                elif \"activitySegment\" in timeline_object:\n",
    "                    activity_segment_json = timeline_object[\"activitySegment\"]      \n",
    "                    \n",
    "                    # # the future columns of our activity_segments dataframe. \n",
    "                        # We identify them thanks to their position in the JSON file.\n",
    "                    activity_segment = {\n",
    "                        \"activityType\": activity_segment_json[\"activityType\"],\n",
    "                        \"startlocation_lat\": activity_segment_json[\"startLocation\"][\"latitudeE7\"],\n",
    "                        \"startlocation_long\": activity_segment_json[\"startLocation\"][\"longitudeE7\"],\n",
    "                        \"endlocation_lat\": activity_segment_json[\"endLocation\"][\"latitudeE7\"],\n",
    "                        \"endlocation_long\": activity_segment_json[\"endLocation\"][\"longitudeE7\"],\n",
    "                        \"startTimestamp\": activity_segment_json[\"duration\"][\"startTimestamp\"],\n",
    "                        \"endTimestamp\": activity_segment_json[\"duration\"][\"endTimestamp\"]\n",
    "                    }\n",
    "                    \n",
    "                    # Optional columns in the timelineObjects object\n",
    "                    for optional_field in [\"distance\", \"confidence\", \"waypointPath\", \"simplifiedRawPath\", \"transitPath\"]:\n",
    "                        if optional_field in activity_segment_json:\n",
    "                            activity_segment[optional_field] = activity_segment_json[optional_field]\n",
    "                        else:\n",
    "                            activity_segment[optional_field] = None\n",
    "                   \n",
    "                # Final step: Concatenate the information from all files into the array activity segments\n",
    "                activity_segments.append(activity_segment)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "751b1270",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'placeId': 'ChIJYbOdsB2I5kcR7rvXL4ZrSnA',\n",
       " 'locationConfidence': 98.007126,\n",
       " 'startTimestamp': '2020-04-19T13:49:48.739Z',\n",
       " 'endTimestamp': '2020-04-20T18:47:02.018Z',\n",
       " 'placeVisitImportance': 'MAIN',\n",
       " 'placeVisitType': 'SINGLE_PLACE',\n",
       " 'latitudeE7': 488745083,\n",
       " 'longitudeE7': 20972042,\n",
       " 'centerLatE7': 488743594,\n",
       " 'centerLngE7': 20972360,\n",
       " 'name': None,\n",
       " 'address': '40 Chem. du Clos Courché, 78160 Marly-le-Roi, France'}"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Glimpse of what the list place_visits contains (e.g. the first element in the list)\n",
    "place_visits[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "642bc21a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'activityType': 'WALKING',\n",
       " 'startlocation_lat': 488740928,\n",
       " 'startlocation_long': 20967533,\n",
       " 'endlocation_lat': 488738457,\n",
       " 'endlocation_long': 20964533,\n",
       " 'startTimestamp': '2020-04-19T13:03:40.171Z',\n",
       " 'endTimestamp': '2020-04-19T13:49:48.739Z',\n",
       " 'distance': 3982,\n",
       " 'confidence': 'HIGH',\n",
       " 'waypointPath': {'waypoints': [{'latE7': 488741226, 'lngE7': 20966773},\n",
       "   {'latE7': 488727912, 'lngE7': 20932619},\n",
       "   {'latE7': 488730621, 'lngE7': 20855600},\n",
       "   {'latE7': 488708419, 'lngE7': 20799911},\n",
       "   {'latE7': 488704833, 'lngE7': 20790579},\n",
       "   {'latE7': 488698654, 'lngE7': 20782673},\n",
       "   {'latE7': 488664665, 'lngE7': 20773117},\n",
       "   {'latE7': 488696136, 'lngE7': 20855243},\n",
       "   {'latE7': 488702735, 'lngE7': 20890526},\n",
       "   {'latE7': 488716583, 'lngE7': 20943946},\n",
       "   {'latE7': 488738517, 'lngE7': 20964334}],\n",
       "  'source': 'INFERRED',\n",
       "  'distanceMeters': 4235.515962182255,\n",
       "  'travelMode': 'WALK',\n",
       "  'confidence': 1.0},\n",
       " 'simplifiedRawPath': {'points': [{'latE7': 488732491,\n",
       "    'lngE7': 20856502,\n",
       "    'accuracyMeters': 14,\n",
       "    'timestamp': '2020-04-19T13:14:53.286Z'},\n",
       "   {'latE7': 488705292,\n",
       "    'lngE7': 20790286,\n",
       "    'accuracyMeters': 13,\n",
       "    'timestamp': '2020-04-19T13:21:06.219Z'},\n",
       "   {'latE7': 488666000,\n",
       "    'lngE7': 20774069,\n",
       "    'accuracyMeters': 26,\n",
       "    'timestamp': '2020-04-19T13:27:41.417Z'},\n",
       "   {'latE7': 488703041,\n",
       "    'lngE7': 20889738,\n",
       "    'accuracyMeters': 15,\n",
       "    'timestamp': '2020-04-19T13:39:36.173Z'}]},\n",
       " 'transitPath': None}"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Glimpse of what the activity_segments contains (e.g. the first element in the list)\n",
    "activity_segments[0]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "90107a7b",
   "metadata": {},
   "source": [
    "<div style=\"background-color: RGB(0,150,250);\" >\n",
    "<h2 style=\"margin: auto; padding: 20px; color:#fff; \">3. Transform the lists into Pandas dataframes</h2> <a class=\"anchor\" id=\"3\"></a>\n",
    "</div>\n",
    "\n",
    "As dataframes are much more convenient than lists for data analysis, we use the pandas library to perform the transformation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "c1ea7a68",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Transform the list into a dataframe\n",
    "place_visits_df = pd.DataFrame(place_visits)\n",
    "activities_df = pd.DataFrame(activity_segments)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "ac80b96c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>placeId</th>\n",
       "      <th>locationConfidence</th>\n",
       "      <th>startTimestamp</th>\n",
       "      <th>endTimestamp</th>\n",
       "      <th>placeVisitImportance</th>\n",
       "      <th>placeVisitType</th>\n",
       "      <th>latitudeE7</th>\n",
       "      <th>longitudeE7</th>\n",
       "      <th>centerLatE7</th>\n",
       "      <th>centerLngE7</th>\n",
       "      <th>name</th>\n",
       "      <th>address</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3081</th>\n",
       "      <td>ChIJ4YbuA1y5FkgRnVJIe4VlqOk</td>\n",
       "      <td>89.39192</td>\n",
       "      <td>2017-08-15T13:54:46.243Z</td>\n",
       "      <td>2017-08-15T15:06:45.530Z</td>\n",
       "      <td>MAIN</td>\n",
       "      <td>SINGLE_PLACE</td>\n",
       "      <td>483862918</td>\n",
       "      <td>-44835494</td>\n",
       "      <td>483864727</td>\n",
       "      <td>-44840673</td>\n",
       "      <td>Square Alphonse Juin</td>\n",
       "      <td>10-, 14 Rue Frégate la Belle Poule, 29200 Bres...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3082</th>\n",
       "      <td>ChIJ0xnzm2G5FkgRSMgfQYsYEpM</td>\n",
       "      <td>99.66684</td>\n",
       "      <td>2017-08-15T15:09:49.427Z</td>\n",
       "      <td>2017-08-15T15:55:54.072Z</td>\n",
       "      <td>MAIN</td>\n",
       "      <td>SINGLE_PLACE</td>\n",
       "      <td>483876587</td>\n",
       "      <td>-44801565</td>\n",
       "      <td>483872117</td>\n",
       "      <td>-44804356</td>\n",
       "      <td>Brest Train Station</td>\n",
       "      <td>08 Place du 19ème Régiment d'Infanterie, 61134...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3080</th>\n",
       "      <td>ChIJz2lWBU3IFkgRe0_6oTJ5Mpw</td>\n",
       "      <td>99.90851</td>\n",
       "      <td>2017-08-10T11:05:21.169Z</td>\n",
       "      <td>2017-08-15T12:39:04.540Z</td>\n",
       "      <td>MAIN</td>\n",
       "      <td>SINGLE_PLACE</td>\n",
       "      <td>482934855</td>\n",
       "      <td>-41780653</td>\n",
       "      <td>482348423</td>\n",
       "      <td>-44958069</td>\n",
       "      <td>Armorique Regional Park</td>\n",
       "      <td>15 Place aux Foires, 29590 Le Faou, France</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                          placeId  locationConfidence  \\\n",
       "3081  ChIJ4YbuA1y5FkgRnVJIe4VlqOk            89.39192   \n",
       "3082  ChIJ0xnzm2G5FkgRSMgfQYsYEpM            99.66684   \n",
       "3080  ChIJz2lWBU3IFkgRe0_6oTJ5Mpw            99.90851   \n",
       "\n",
       "                startTimestamp              endTimestamp placeVisitImportance  \\\n",
       "3081  2017-08-15T13:54:46.243Z  2017-08-15T15:06:45.530Z                 MAIN   \n",
       "3082  2017-08-15T15:09:49.427Z  2017-08-15T15:55:54.072Z                 MAIN   \n",
       "3080  2017-08-10T11:05:21.169Z  2017-08-15T12:39:04.540Z                 MAIN   \n",
       "\n",
       "     placeVisitType  latitudeE7  longitudeE7  centerLatE7  centerLngE7  \\\n",
       "3081   SINGLE_PLACE   483862918    -44835494    483864727    -44840673   \n",
       "3082   SINGLE_PLACE   483876587    -44801565    483872117    -44804356   \n",
       "3080   SINGLE_PLACE   482934855    -41780653    482348423    -44958069   \n",
       "\n",
       "                         name  \\\n",
       "3081     Square Alphonse Juin   \n",
       "3082      Brest Train Station   \n",
       "3080  Armorique Regional Park   \n",
       "\n",
       "                                                address  \n",
       "3081  10-, 14 Rue Frégate la Belle Poule, 29200 Bres...  \n",
       "3082  08 Place du 19ème Régiment d'Infanterie, 61134...  \n",
       "3080         15 Place aux Foires, 29590 Le Faou, France  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Overview of the place_visits dataframe\n",
    "    # For privacy reasons, I display old data\n",
    "place_visits_df.sort_values(by= \"longitudeE7\").head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "23ad2c69",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>activityType</th>\n",
       "      <th>startlocation_lat</th>\n",
       "      <th>startlocation_long</th>\n",
       "      <th>endlocation_lat</th>\n",
       "      <th>endlocation_long</th>\n",
       "      <th>startTimestamp</th>\n",
       "      <th>endTimestamp</th>\n",
       "      <th>distance</th>\n",
       "      <th>confidence</th>\n",
       "      <th>waypointPath</th>\n",
       "      <th>simplifiedRawPath</th>\n",
       "      <th>transitPath</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>7014</th>\n",
       "      <td>RUNNING</td>\n",
       "      <td>482347212</td>\n",
       "      <td>-44959911</td>\n",
       "      <td>482934855</td>\n",
       "      <td>-41780653</td>\n",
       "      <td>2017-08-09T14:11:00.922Z</td>\n",
       "      <td>2017-08-09T14:49:30.632Z</td>\n",
       "      <td>3463.0</td>\n",
       "      <td>MEDIUM</td>\n",
       "      <td>None</td>\n",
       "      <td>{'points': [{'latE7': 482264144, 'lngE7': -450...</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7015</th>\n",
       "      <td>RUNNING</td>\n",
       "      <td>482347212</td>\n",
       "      <td>-44959911</td>\n",
       "      <td>482934855</td>\n",
       "      <td>-41780653</td>\n",
       "      <td>2017-08-09T14:11:00.922Z</td>\n",
       "      <td>2017-08-09T14:49:30.632Z</td>\n",
       "      <td>3463.0</td>\n",
       "      <td>MEDIUM</td>\n",
       "      <td>None</td>\n",
       "      <td>{'points': [{'latE7': 482264144, 'lngE7': -450...</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7013</th>\n",
       "      <td>IN_PASSENGER_VEHICLE</td>\n",
       "      <td>482348022</td>\n",
       "      <td>-44959138</td>\n",
       "      <td>482934855</td>\n",
       "      <td>-41780653</td>\n",
       "      <td>2017-08-05T17:19:04.582Z</td>\n",
       "      <td>2017-08-05T18:10:18.126Z</td>\n",
       "      <td>985989.0</td>\n",
       "      <td>HIGH</td>\n",
       "      <td>{'waypoints': [{'latE7': 482347145, 'lngE7': -...</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              activityType  startlocation_lat  startlocation_long  \\\n",
       "7014               RUNNING          482347212           -44959911   \n",
       "7015               RUNNING          482347212           -44959911   \n",
       "7013  IN_PASSENGER_VEHICLE          482348022           -44959138   \n",
       "\n",
       "      endlocation_lat  endlocation_long            startTimestamp  \\\n",
       "7014        482934855         -41780653  2017-08-09T14:11:00.922Z   \n",
       "7015        482934855         -41780653  2017-08-09T14:11:00.922Z   \n",
       "7013        482934855         -41780653  2017-08-05T17:19:04.582Z   \n",
       "\n",
       "                  endTimestamp  distance confidence  \\\n",
       "7014  2017-08-09T14:49:30.632Z    3463.0     MEDIUM   \n",
       "7015  2017-08-09T14:49:30.632Z    3463.0     MEDIUM   \n",
       "7013  2017-08-05T18:10:18.126Z  985989.0       HIGH   \n",
       "\n",
       "                                           waypointPath  \\\n",
       "7014                                               None   \n",
       "7015                                               None   \n",
       "7013  {'waypoints': [{'latE7': 482347145, 'lngE7': -...   \n",
       "\n",
       "                                      simplifiedRawPath transitPath  \n",
       "7014  {'points': [{'latE7': 482264144, 'lngE7': -450...        None  \n",
       "7015  {'points': [{'latE7': 482264144, 'lngE7': -450...        None  \n",
       "7013                                               None        None  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Overview of the activities dataframe\n",
    "    # For privacy reasons, I display old data\n",
    "activities_df.sort_values(\"startlocation_long\").head(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "52e15148",
   "metadata": {},
   "source": [
    "<div style=\"background-color: RGB(0,150,250);\" >\n",
    "<h2 style=\"margin: auto; padding: 20px; color:#fff; \">4. Data cleaning</h2> <a class=\"anchor\" id=\"4\"></a>\n",
    "</div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "3b188618",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Cleaning 1: Convert timestamp variables to datetime types\n",
    "place_visits_df[[\"startTimestamp\", \"endTimestamp\"]] = place_visits_df[[\"startTimestamp\", \"endTimestamp\"]].apply(pd.to_datetime)\n",
    "activities_df[[\"startTimestamp\", \"endTimestamp\"]] = activities_df[[\"startTimestamp\", \"endTimestamp\"]].apply(pd.to_datetime)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "f1833725",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Cleaning 2: Obtain the latitude and longitude in the proper format (floats)\n",
    "\n",
    "# Step 1: Create a function to divide the Google latitude and longitude E7 by 10,000,000\n",
    "def f(x): \n",
    "    return x/1E7\n",
    "\n",
    "# Step 2: Apply the function to all geo coordinates variables\n",
    "place_visits_df[[\"latitude\", \"longitude\",\n",
    "                 \"centerLat\", \"centerLng\"]] = place_visits_df[[\"latitudeE7\", \"longitudeE7\",\n",
    "                                                               \"centerLatE7\", \"centerLngE7\"]].apply(f, axis = 1)\n",
    "\n",
    "activities_df[[\"startlocation_lat\", \"startlocation_long\", \n",
    "               \"endlocation_lat\", \"endlocation_long\"]] = activities_df[[\"startlocation_lat\", \"startlocation_long\", \n",
    "               \"endlocation_lat\", \"endlocation_long\"]].apply(f, axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "20df0d79",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>placeId</th>\n",
       "      <th>locationConfidence</th>\n",
       "      <th>startTimestamp</th>\n",
       "      <th>endTimestamp</th>\n",
       "      <th>placeVisitImportance</th>\n",
       "      <th>placeVisitType</th>\n",
       "      <th>latitudeE7</th>\n",
       "      <th>longitudeE7</th>\n",
       "      <th>centerLatE7</th>\n",
       "      <th>centerLngE7</th>\n",
       "      <th>name</th>\n",
       "      <th>address</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>centerLat</th>\n",
       "      <th>centerLng</th>\n",
       "      <th>address_clean</th>\n",
       "      <th>name_clean</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3081</th>\n",
       "      <td>ChIJ4YbuA1y5FkgRnVJIe4VlqOk</td>\n",
       "      <td>89.39192</td>\n",
       "      <td>2017-08-15 13:54:46.243000+00:00</td>\n",
       "      <td>2017-08-15 15:06:45.530000+00:00</td>\n",
       "      <td>MAIN</td>\n",
       "      <td>SINGLE_PLACE</td>\n",
       "      <td>483862918</td>\n",
       "      <td>-44835494</td>\n",
       "      <td>483864727</td>\n",
       "      <td>-44840673</td>\n",
       "      <td>Square Alphonse Juin</td>\n",
       "      <td>10-, 14 Rue Frégate la Belle Poule, 29200 Bres...</td>\n",
       "      <td>48.386292</td>\n",
       "      <td>-4.483549</td>\n",
       "      <td>48.386473</td>\n",
       "      <td>-4.484067</td>\n",
       "      <td>10-, 14 RUE FREGATE LA BELLE POULE, 29200 BRES...</td>\n",
       "      <td>SQUARE ALPHONSE JUIN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3082</th>\n",
       "      <td>ChIJ0xnzm2G5FkgRSMgfQYsYEpM</td>\n",
       "      <td>99.66684</td>\n",
       "      <td>2017-08-15 15:09:49.427000+00:00</td>\n",
       "      <td>2017-08-15 15:55:54.072000+00:00</td>\n",
       "      <td>MAIN</td>\n",
       "      <td>SINGLE_PLACE</td>\n",
       "      <td>483876587</td>\n",
       "      <td>-44801565</td>\n",
       "      <td>483872117</td>\n",
       "      <td>-44804356</td>\n",
       "      <td>Brest Train Station</td>\n",
       "      <td>08 Place du 19ème Régiment d'Infanterie, 61134...</td>\n",
       "      <td>48.387659</td>\n",
       "      <td>-4.480156</td>\n",
       "      <td>48.387212</td>\n",
       "      <td>-4.480436</td>\n",
       "      <td>08 PLACE DU 19EME REGIMENT D'INFANTERIE, 61134...</td>\n",
       "      <td>BREST TRAIN STATION</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3080</th>\n",
       "      <td>ChIJz2lWBU3IFkgRe0_6oTJ5Mpw</td>\n",
       "      <td>99.90851</td>\n",
       "      <td>2017-08-10 11:05:21.169000+00:00</td>\n",
       "      <td>2017-08-15 12:39:04.540000+00:00</td>\n",
       "      <td>MAIN</td>\n",
       "      <td>SINGLE_PLACE</td>\n",
       "      <td>482934855</td>\n",
       "      <td>-41780653</td>\n",
       "      <td>482348423</td>\n",
       "      <td>-44958069</td>\n",
       "      <td>Armorique Regional Park</td>\n",
       "      <td>15 Place aux Foires, 29590 Le Faou, France</td>\n",
       "      <td>48.293486</td>\n",
       "      <td>-4.178065</td>\n",
       "      <td>48.234842</td>\n",
       "      <td>-4.495807</td>\n",
       "      <td>15 PLACE AUX FOIRES, 29590 LE FAOU, FRANCE</td>\n",
       "      <td>ARMORIQUE REGIONAL PARK</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                          placeId  locationConfidence  \\\n",
       "3081  ChIJ4YbuA1y5FkgRnVJIe4VlqOk            89.39192   \n",
       "3082  ChIJ0xnzm2G5FkgRSMgfQYsYEpM            99.66684   \n",
       "3080  ChIJz2lWBU3IFkgRe0_6oTJ5Mpw            99.90851   \n",
       "\n",
       "                       startTimestamp                     endTimestamp  \\\n",
       "3081 2017-08-15 13:54:46.243000+00:00 2017-08-15 15:06:45.530000+00:00   \n",
       "3082 2017-08-15 15:09:49.427000+00:00 2017-08-15 15:55:54.072000+00:00   \n",
       "3080 2017-08-10 11:05:21.169000+00:00 2017-08-15 12:39:04.540000+00:00   \n",
       "\n",
       "     placeVisitImportance placeVisitType  latitudeE7  longitudeE7  \\\n",
       "3081                 MAIN   SINGLE_PLACE   483862918    -44835494   \n",
       "3082                 MAIN   SINGLE_PLACE   483876587    -44801565   \n",
       "3080                 MAIN   SINGLE_PLACE   482934855    -41780653   \n",
       "\n",
       "      centerLatE7  centerLngE7                     name  \\\n",
       "3081    483864727    -44840673     Square Alphonse Juin   \n",
       "3082    483872117    -44804356      Brest Train Station   \n",
       "3080    482348423    -44958069  Armorique Regional Park   \n",
       "\n",
       "                                                address   latitude  longitude  \\\n",
       "3081  10-, 14 Rue Frégate la Belle Poule, 29200 Bres...  48.386292  -4.483549   \n",
       "3082  08 Place du 19ème Régiment d'Infanterie, 61134...  48.387659  -4.480156   \n",
       "3080         15 Place aux Foires, 29590 Le Faou, France  48.293486  -4.178065   \n",
       "\n",
       "      centerLat  centerLng                                      address_clean  \\\n",
       "3081  48.386473  -4.484067  10-, 14 RUE FREGATE LA BELLE POULE, 29200 BRES...   \n",
       "3082  48.387212  -4.480436  08 PLACE DU 19EME REGIMENT D'INFANTERIE, 61134...   \n",
       "3080  48.234842  -4.495807         15 PLACE AUX FOIRES, 29590 LE FAOU, FRANCE   \n",
       "\n",
       "                   name_clean  \n",
       "3081     SQUARE ALPHONSE JUIN  \n",
       "3082      BREST TRAIN STATION  \n",
       "3080  ARMORIQUE REGIONAL PARK  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Cleaning 3: Create clean address and name variables with no French accents and written in capital letters\n",
    "place_visits_df[\"address_clean\"] = place_visits_df[\"address\"].str.normalize('NFKD').str.encode('ascii', \n",
    "                                                                                         errors='ignore').str.decode('utf-8').str.upper()\n",
    "place_visits_df['address_clean'] = place_visits_df['address_clean'].str.replace('ROAD','RD')\n",
    "\n",
    "place_visits_df[\"name_clean\"] = place_visits_df[\"name\"].str.normalize('NFKD').str.encode('ascii', \n",
    "                                                                                         errors='ignore').str.decode('utf-8').str.upper()\n",
    "place_visits_df.sort_values(by = \"longitudeE7\").head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "4b7b8ba8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>placeId</th>\n",
       "      <th>locationConfidence</th>\n",
       "      <th>startTimestamp</th>\n",
       "      <th>endTimestamp</th>\n",
       "      <th>placeVisitImportance</th>\n",
       "      <th>placeVisitType</th>\n",
       "      <th>name</th>\n",
       "      <th>address</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>centerLat</th>\n",
       "      <th>centerLng</th>\n",
       "      <th>address_clean</th>\n",
       "      <th>name_clean</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3081</th>\n",
       "      <td>ChIJ4YbuA1y5FkgRnVJIe4VlqOk</td>\n",
       "      <td>89.39192</td>\n",
       "      <td>2017-08-15 13:54:46.243000+00:00</td>\n",
       "      <td>2017-08-15 15:06:45.530000+00:00</td>\n",
       "      <td>MAIN</td>\n",
       "      <td>SINGLE_PLACE</td>\n",
       "      <td>Square Alphonse Juin</td>\n",
       "      <td>10-, 14 Rue Frégate la Belle Poule, 29200 Bres...</td>\n",
       "      <td>48.386292</td>\n",
       "      <td>-4.483549</td>\n",
       "      <td>48.386473</td>\n",
       "      <td>-4.484067</td>\n",
       "      <td>10-, 14 RUE FREGATE LA BELLE POULE, 29200 BRES...</td>\n",
       "      <td>SQUARE ALPHONSE JUIN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3082</th>\n",
       "      <td>ChIJ0xnzm2G5FkgRSMgfQYsYEpM</td>\n",
       "      <td>99.66684</td>\n",
       "      <td>2017-08-15 15:09:49.427000+00:00</td>\n",
       "      <td>2017-08-15 15:55:54.072000+00:00</td>\n",
       "      <td>MAIN</td>\n",
       "      <td>SINGLE_PLACE</td>\n",
       "      <td>Brest Train Station</td>\n",
       "      <td>08 Place du 19ème Régiment d'Infanterie, 61134...</td>\n",
       "      <td>48.387659</td>\n",
       "      <td>-4.480156</td>\n",
       "      <td>48.387212</td>\n",
       "      <td>-4.480436</td>\n",
       "      <td>08 PLACE DU 19EME REGIMENT D'INFANTERIE, 61134...</td>\n",
       "      <td>BREST TRAIN STATION</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3080</th>\n",
       "      <td>ChIJz2lWBU3IFkgRe0_6oTJ5Mpw</td>\n",
       "      <td>99.90851</td>\n",
       "      <td>2017-08-10 11:05:21.169000+00:00</td>\n",
       "      <td>2017-08-15 12:39:04.540000+00:00</td>\n",
       "      <td>MAIN</td>\n",
       "      <td>SINGLE_PLACE</td>\n",
       "      <td>Armorique Regional Park</td>\n",
       "      <td>15 Place aux Foires, 29590 Le Faou, France</td>\n",
       "      <td>48.293486</td>\n",
       "      <td>-4.178065</td>\n",
       "      <td>48.234842</td>\n",
       "      <td>-4.495807</td>\n",
       "      <td>15 PLACE AUX FOIRES, 29590 LE FAOU, FRANCE</td>\n",
       "      <td>ARMORIQUE REGIONAL PARK</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                          placeId  locationConfidence  \\\n",
       "3081  ChIJ4YbuA1y5FkgRnVJIe4VlqOk            89.39192   \n",
       "3082  ChIJ0xnzm2G5FkgRSMgfQYsYEpM            99.66684   \n",
       "3080  ChIJz2lWBU3IFkgRe0_6oTJ5Mpw            99.90851   \n",
       "\n",
       "                       startTimestamp                     endTimestamp  \\\n",
       "3081 2017-08-15 13:54:46.243000+00:00 2017-08-15 15:06:45.530000+00:00   \n",
       "3082 2017-08-15 15:09:49.427000+00:00 2017-08-15 15:55:54.072000+00:00   \n",
       "3080 2017-08-10 11:05:21.169000+00:00 2017-08-15 12:39:04.540000+00:00   \n",
       "\n",
       "     placeVisitImportance placeVisitType                     name  \\\n",
       "3081                 MAIN   SINGLE_PLACE     Square Alphonse Juin   \n",
       "3082                 MAIN   SINGLE_PLACE      Brest Train Station   \n",
       "3080                 MAIN   SINGLE_PLACE  Armorique Regional Park   \n",
       "\n",
       "                                                address   latitude  longitude  \\\n",
       "3081  10-, 14 Rue Frégate la Belle Poule, 29200 Bres...  48.386292  -4.483549   \n",
       "3082  08 Place du 19ème Régiment d'Infanterie, 61134...  48.387659  -4.480156   \n",
       "3080         15 Place aux Foires, 29590 Le Faou, France  48.293486  -4.178065   \n",
       "\n",
       "      centerLat  centerLng                                      address_clean  \\\n",
       "3081  48.386473  -4.484067  10-, 14 RUE FREGATE LA BELLE POULE, 29200 BRES...   \n",
       "3082  48.387212  -4.480436  08 PLACE DU 19EME REGIMENT D'INFANTERIE, 61134...   \n",
       "3080  48.234842  -4.495807         15 PLACE AUX FOIRES, 29590 LE FAOU, FRANCE   \n",
       "\n",
       "                   name_clean  \n",
       "3081     SQUARE ALPHONSE JUIN  \n",
       "3082      BREST TRAIN STATION  \n",
       "3080  ARMORIQUE REGIONAL PARK  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Cleaning 4: Drop variables that are of no use\n",
    "place_visits_df = place_visits_df.drop([\"latitudeE7\", \"longitudeE7\", \"centerLatE7\", \"centerLngE7\"], axis =1)\n",
    "place_visits_df.sort_values(by = \"longitude\").head(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9327df0f",
   "metadata": {},
   "source": [
    "<div style=\"background-color: RGB(0,150,250);\" >\n",
    "<h2 style=\"margin: auto; padding: 20px; color:#fff; \">5. Data enrichment</h2> <a class=\"anchor\" id=\"5\"></a>\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "47898cae",
   "metadata": {},
   "source": [
    "### 5.1. Dates and durations <a class=\"anchor\" id=\"5.1\"></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "94e654fb",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Compute durations for activities and travels\n",
    "place_visits_df[\"duration\"] = place_visits_df.endTimestamp - place_visits_df.startTimestamp\n",
    "place_visits_df[\"duration_minutes\"] = place_visits_df.duration.dt.total_seconds()/60\n",
    "\n",
    "activities_df[\"duration\"]= activities_df.endTimestamp - activities_df.startTimestamp\n",
    "activities_df[\"duration_minutes\"] = activities_df.duration.dt.total_seconds()/60"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "d89fbcb8",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Add year, month and date variables in the place_visits dataframe\n",
    "place_visits_df[\"Year\"] = place_visits_df[\"startTimestamp\"].dt.year\n",
    "place_visits_df[\"Month\"] = place_visits_df[\"startTimestamp\"].dt.month\n",
    "place_visits_df[\"Date\"] = place_visits_df[\"startTimestamp\"].dt.date"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ef4c21cd",
   "metadata": {},
   "source": [
    "### 5.2. Add manually names to addresses with no Google Maps name associated <a class=\"anchor\" id=\"5.2\"></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "dd4dd6a0",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Identify the addresses that do not have any name associated\n",
    "no_name = place_visits_df[place_visits_df['name'].isna()]\n",
    "test = no_name['address'].value_counts(ascending = False).rename_axis('address').to_frame(\"counts\")\n",
    "#test -- I do not display the results for privacy reasons"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "d3d84377",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>placeId</th>\n",
       "      <th>locationConfidence</th>\n",
       "      <th>startTimestamp</th>\n",
       "      <th>endTimestamp</th>\n",
       "      <th>placeVisitImportance</th>\n",
       "      <th>placeVisitType</th>\n",
       "      <th>name</th>\n",
       "      <th>address</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>centerLat</th>\n",
       "      <th>centerLng</th>\n",
       "      <th>address_clean</th>\n",
       "      <th>name_clean</th>\n",
       "      <th>duration</th>\n",
       "      <th>duration_minutes</th>\n",
       "      <th>Year</th>\n",
       "      <th>Month</th>\n",
       "      <th>Date</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3081</th>\n",
       "      <td>ChIJ4YbuA1y5FkgRnVJIe4VlqOk</td>\n",
       "      <td>89.39192</td>\n",
       "      <td>2017-08-15 13:54:46.243000+00:00</td>\n",
       "      <td>2017-08-15 15:06:45.530000+00:00</td>\n",
       "      <td>MAIN</td>\n",
       "      <td>SINGLE_PLACE</td>\n",
       "      <td>Square Alphonse Juin</td>\n",
       "      <td>10-, 14 Rue Frégate la Belle Poule, 29200 Bres...</td>\n",
       "      <td>48.386292</td>\n",
       "      <td>-4.483549</td>\n",
       "      <td>48.386473</td>\n",
       "      <td>-4.484067</td>\n",
       "      <td>10-, 14 RUE FREGATE LA BELLE POULE, 29200 BRES...</td>\n",
       "      <td>SQUARE ALPHONSE JUIN</td>\n",
       "      <td>0 days 01:11:59.287000</td>\n",
       "      <td>71.988117</td>\n",
       "      <td>2017</td>\n",
       "      <td>8</td>\n",
       "      <td>2017-08-15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3082</th>\n",
       "      <td>ChIJ0xnzm2G5FkgRSMgfQYsYEpM</td>\n",
       "      <td>99.66684</td>\n",
       "      <td>2017-08-15 15:09:49.427000+00:00</td>\n",
       "      <td>2017-08-15 15:55:54.072000+00:00</td>\n",
       "      <td>MAIN</td>\n",
       "      <td>SINGLE_PLACE</td>\n",
       "      <td>Brest Train Station</td>\n",
       "      <td>08 Place du 19ème Régiment d'Infanterie, 61134...</td>\n",
       "      <td>48.387659</td>\n",
       "      <td>-4.480156</td>\n",
       "      <td>48.387212</td>\n",
       "      <td>-4.480436</td>\n",
       "      <td>08 PLACE DU 19EME REGIMENT D'INFANTERIE, 61134...</td>\n",
       "      <td>BREST TRAIN STATION</td>\n",
       "      <td>0 days 00:46:04.645000</td>\n",
       "      <td>46.077417</td>\n",
       "      <td>2017</td>\n",
       "      <td>8</td>\n",
       "      <td>2017-08-15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3080</th>\n",
       "      <td>ChIJz2lWBU3IFkgRe0_6oTJ5Mpw</td>\n",
       "      <td>99.90851</td>\n",
       "      <td>2017-08-10 11:05:21.169000+00:00</td>\n",
       "      <td>2017-08-15 12:39:04.540000+00:00</td>\n",
       "      <td>MAIN</td>\n",
       "      <td>SINGLE_PLACE</td>\n",
       "      <td>Armorique Regional Park</td>\n",
       "      <td>15 Place aux Foires, 29590 Le Faou, France</td>\n",
       "      <td>48.293486</td>\n",
       "      <td>-4.178065</td>\n",
       "      <td>48.234842</td>\n",
       "      <td>-4.495807</td>\n",
       "      <td>15 PLACE AUX FOIRES, 29590 LE FAOU, FRANCE</td>\n",
       "      <td>ARMORIQUE REGIONAL PARK</td>\n",
       "      <td>5 days 01:33:43.371000</td>\n",
       "      <td>7293.722850</td>\n",
       "      <td>2017</td>\n",
       "      <td>8</td>\n",
       "      <td>2017-08-10</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                          placeId  locationConfidence  \\\n",
       "3081  ChIJ4YbuA1y5FkgRnVJIe4VlqOk            89.39192   \n",
       "3082  ChIJ0xnzm2G5FkgRSMgfQYsYEpM            99.66684   \n",
       "3080  ChIJz2lWBU3IFkgRe0_6oTJ5Mpw            99.90851   \n",
       "\n",
       "                       startTimestamp                     endTimestamp  \\\n",
       "3081 2017-08-15 13:54:46.243000+00:00 2017-08-15 15:06:45.530000+00:00   \n",
       "3082 2017-08-15 15:09:49.427000+00:00 2017-08-15 15:55:54.072000+00:00   \n",
       "3080 2017-08-10 11:05:21.169000+00:00 2017-08-15 12:39:04.540000+00:00   \n",
       "\n",
       "     placeVisitImportance placeVisitType                     name  \\\n",
       "3081                 MAIN   SINGLE_PLACE     Square Alphonse Juin   \n",
       "3082                 MAIN   SINGLE_PLACE      Brest Train Station   \n",
       "3080                 MAIN   SINGLE_PLACE  Armorique Regional Park   \n",
       "\n",
       "                                                address   latitude  longitude  \\\n",
       "3081  10-, 14 Rue Frégate la Belle Poule, 29200 Bres...  48.386292  -4.483549   \n",
       "3082  08 Place du 19ème Régiment d'Infanterie, 61134...  48.387659  -4.480156   \n",
       "3080         15 Place aux Foires, 29590 Le Faou, France  48.293486  -4.178065   \n",
       "\n",
       "      centerLat  centerLng                                      address_clean  \\\n",
       "3081  48.386473  -4.484067  10-, 14 RUE FREGATE LA BELLE POULE, 29200 BRES...   \n",
       "3082  48.387212  -4.480436  08 PLACE DU 19EME REGIMENT D'INFANTERIE, 61134...   \n",
       "3080  48.234842  -4.495807         15 PLACE AUX FOIRES, 29590 LE FAOU, FRANCE   \n",
       "\n",
       "                   name_clean               duration  duration_minutes  Year  \\\n",
       "3081     SQUARE ALPHONSE JUIN 0 days 01:11:59.287000         71.988117  2017   \n",
       "3082      BREST TRAIN STATION 0 days 00:46:04.645000         46.077417  2017   \n",
       "3080  ARMORIQUE REGIONAL PARK 5 days 01:33:43.371000       7293.722850  2017   \n",
       "\n",
       "      Month        Date  \n",
       "3081      8  2017-08-15  \n",
       "3082      8  2017-08-15  \n",
       "3080      8  2017-08-10  "
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Give a name to the most popular addresses that have no name associated\n",
    "\n",
    "# Jupyter notebook public version: I do not display the code for confidentiality reasons.\n",
    "# But basically, I created 2 variables :\n",
    "    # name_clean : I attribute names to the addresses I lived and worked, and that Google did not recognize\n",
    "    # Address_type : 'HOME' or 'WORK' or Null values\n",
    "place_visits_df.sort_values(\"longitude\").head(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d7e0e56a",
   "metadata": {},
   "source": [
    "### 5.3. Capture City, Zipcode and Country informations <a class=\"anchor\" id=\"5.3\"></a>\n",
    "\n",
    "The method used below is a bit \"quick and dirty\" but that is the best I came up with so far."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "5bef2c26",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>placeId</th>\n",
       "      <th>locationConfidence</th>\n",
       "      <th>startTimestamp</th>\n",
       "      <th>endTimestamp</th>\n",
       "      <th>placeVisitImportance</th>\n",
       "      <th>placeVisitType</th>\n",
       "      <th>name</th>\n",
       "      <th>address</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>centerLat</th>\n",
       "      <th>centerLng</th>\n",
       "      <th>address_clean</th>\n",
       "      <th>name_clean</th>\n",
       "      <th>duration</th>\n",
       "      <th>duration_minutes</th>\n",
       "      <th>Year</th>\n",
       "      <th>Month</th>\n",
       "      <th>Date</th>\n",
       "      <th>nb_commas</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3081</th>\n",
       "      <td>ChIJ4YbuA1y5FkgRnVJIe4VlqOk</td>\n",
       "      <td>89.39192</td>\n",
       "      <td>2017-08-15 13:54:46.243000+00:00</td>\n",
       "      <td>2017-08-15 15:06:45.530000+00:00</td>\n",
       "      <td>MAIN</td>\n",
       "      <td>SINGLE_PLACE</td>\n",
       "      <td>Square Alphonse Juin</td>\n",
       "      <td>10-, 14 Rue Frégate la Belle Poule, 29200 Bres...</td>\n",
       "      <td>48.386292</td>\n",
       "      <td>-4.483549</td>\n",
       "      <td>48.386473</td>\n",
       "      <td>-4.484067</td>\n",
       "      <td>10-, 14 RUE FREGATE LA BELLE POULE, 29200 BRES...</td>\n",
       "      <td>SQUARE ALPHONSE JUIN</td>\n",
       "      <td>0 days 01:11:59.287000</td>\n",
       "      <td>71.988117</td>\n",
       "      <td>2017</td>\n",
       "      <td>8</td>\n",
       "      <td>2017-08-15</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3082</th>\n",
       "      <td>ChIJ0xnzm2G5FkgRSMgfQYsYEpM</td>\n",
       "      <td>99.66684</td>\n",
       "      <td>2017-08-15 15:09:49.427000+00:00</td>\n",
       "      <td>2017-08-15 15:55:54.072000+00:00</td>\n",
       "      <td>MAIN</td>\n",
       "      <td>SINGLE_PLACE</td>\n",
       "      <td>Brest Train Station</td>\n",
       "      <td>08 Place du 19ème Régiment d'Infanterie, 61134...</td>\n",
       "      <td>48.387659</td>\n",
       "      <td>-4.480156</td>\n",
       "      <td>48.387212</td>\n",
       "      <td>-4.480436</td>\n",
       "      <td>08 PLACE DU 19EME REGIMENT D'INFANTERIE, 61134...</td>\n",
       "      <td>BREST TRAIN STATION</td>\n",
       "      <td>0 days 00:46:04.645000</td>\n",
       "      <td>46.077417</td>\n",
       "      <td>2017</td>\n",
       "      <td>8</td>\n",
       "      <td>2017-08-15</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3080</th>\n",
       "      <td>ChIJz2lWBU3IFkgRe0_6oTJ5Mpw</td>\n",
       "      <td>99.90851</td>\n",
       "      <td>2017-08-10 11:05:21.169000+00:00</td>\n",
       "      <td>2017-08-15 12:39:04.540000+00:00</td>\n",
       "      <td>MAIN</td>\n",
       "      <td>SINGLE_PLACE</td>\n",
       "      <td>Armorique Regional Park</td>\n",
       "      <td>15 Place aux Foires, 29590 Le Faou, France</td>\n",
       "      <td>48.293486</td>\n",
       "      <td>-4.178065</td>\n",
       "      <td>48.234842</td>\n",
       "      <td>-4.495807</td>\n",
       "      <td>15 PLACE AUX FOIRES, 29590 LE FAOU, FRANCE</td>\n",
       "      <td>ARMORIQUE REGIONAL PARK</td>\n",
       "      <td>5 days 01:33:43.371000</td>\n",
       "      <td>7293.722850</td>\n",
       "      <td>2017</td>\n",
       "      <td>8</td>\n",
       "      <td>2017-08-10</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3077</th>\n",
       "      <td>ChIJz2lWBU3IFkgRe0_6oTJ5Mpw</td>\n",
       "      <td>99.88017</td>\n",
       "      <td>2017-08-05 13:42:39.180000+00:00</td>\n",
       "      <td>2017-08-05 17:19:04.582000+00:00</td>\n",
       "      <td>MAIN</td>\n",
       "      <td>SINGLE_PLACE</td>\n",
       "      <td>Armorique Regional Park</td>\n",
       "      <td>15 Place aux Foires, 29590 Le Faou, France</td>\n",
       "      <td>48.293486</td>\n",
       "      <td>-4.178065</td>\n",
       "      <td>48.234830</td>\n",
       "      <td>-4.495876</td>\n",
       "      <td>15 PLACE AUX FOIRES, 29590 LE FAOU, FRANCE</td>\n",
       "      <td>ARMORIQUE REGIONAL PARK</td>\n",
       "      <td>0 days 03:36:25.402000</td>\n",
       "      <td>216.423367</td>\n",
       "      <td>2017</td>\n",
       "      <td>8</td>\n",
       "      <td>2017-08-05</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3078</th>\n",
       "      <td>ChIJz2lWBU3IFkgRe0_6oTJ5Mpw</td>\n",
       "      <td>99.94573</td>\n",
       "      <td>2017-08-05 18:10:18.126000+00:00</td>\n",
       "      <td>2017-08-09 14:11:00.922000+00:00</td>\n",
       "      <td>MAIN</td>\n",
       "      <td>SINGLE_PLACE</td>\n",
       "      <td>Armorique Regional Park</td>\n",
       "      <td>15 Place aux Foires, 29590 Le Faou, France</td>\n",
       "      <td>48.293486</td>\n",
       "      <td>-4.178065</td>\n",
       "      <td>48.234834</td>\n",
       "      <td>-4.495802</td>\n",
       "      <td>15 PLACE AUX FOIRES, 29590 LE FAOU, FRANCE</td>\n",
       "      <td>ARMORIQUE REGIONAL PARK</td>\n",
       "      <td>3 days 20:00:42.796000</td>\n",
       "      <td>5520.713267</td>\n",
       "      <td>2017</td>\n",
       "      <td>8</td>\n",
       "      <td>2017-08-05</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                          placeId  locationConfidence  \\\n",
       "3081  ChIJ4YbuA1y5FkgRnVJIe4VlqOk            89.39192   \n",
       "3082  ChIJ0xnzm2G5FkgRSMgfQYsYEpM            99.66684   \n",
       "3080  ChIJz2lWBU3IFkgRe0_6oTJ5Mpw            99.90851   \n",
       "3077  ChIJz2lWBU3IFkgRe0_6oTJ5Mpw            99.88017   \n",
       "3078  ChIJz2lWBU3IFkgRe0_6oTJ5Mpw            99.94573   \n",
       "\n",
       "                       startTimestamp                     endTimestamp  \\\n",
       "3081 2017-08-15 13:54:46.243000+00:00 2017-08-15 15:06:45.530000+00:00   \n",
       "3082 2017-08-15 15:09:49.427000+00:00 2017-08-15 15:55:54.072000+00:00   \n",
       "3080 2017-08-10 11:05:21.169000+00:00 2017-08-15 12:39:04.540000+00:00   \n",
       "3077 2017-08-05 13:42:39.180000+00:00 2017-08-05 17:19:04.582000+00:00   \n",
       "3078 2017-08-05 18:10:18.126000+00:00 2017-08-09 14:11:00.922000+00:00   \n",
       "\n",
       "     placeVisitImportance placeVisitType                     name  \\\n",
       "3081                 MAIN   SINGLE_PLACE     Square Alphonse Juin   \n",
       "3082                 MAIN   SINGLE_PLACE      Brest Train Station   \n",
       "3080                 MAIN   SINGLE_PLACE  Armorique Regional Park   \n",
       "3077                 MAIN   SINGLE_PLACE  Armorique Regional Park   \n",
       "3078                 MAIN   SINGLE_PLACE  Armorique Regional Park   \n",
       "\n",
       "                                                address   latitude  longitude  \\\n",
       "3081  10-, 14 Rue Frégate la Belle Poule, 29200 Bres...  48.386292  -4.483549   \n",
       "3082  08 Place du 19ème Régiment d'Infanterie, 61134...  48.387659  -4.480156   \n",
       "3080         15 Place aux Foires, 29590 Le Faou, France  48.293486  -4.178065   \n",
       "3077         15 Place aux Foires, 29590 Le Faou, France  48.293486  -4.178065   \n",
       "3078         15 Place aux Foires, 29590 Le Faou, France  48.293486  -4.178065   \n",
       "\n",
       "      centerLat  centerLng                                      address_clean  \\\n",
       "3081  48.386473  -4.484067  10-, 14 RUE FREGATE LA BELLE POULE, 29200 BRES...   \n",
       "3082  48.387212  -4.480436  08 PLACE DU 19EME REGIMENT D'INFANTERIE, 61134...   \n",
       "3080  48.234842  -4.495807         15 PLACE AUX FOIRES, 29590 LE FAOU, FRANCE   \n",
       "3077  48.234830  -4.495876         15 PLACE AUX FOIRES, 29590 LE FAOU, FRANCE   \n",
       "3078  48.234834  -4.495802         15 PLACE AUX FOIRES, 29590 LE FAOU, FRANCE   \n",
       "\n",
       "                   name_clean               duration  duration_minutes  Year  \\\n",
       "3081     SQUARE ALPHONSE JUIN 0 days 01:11:59.287000         71.988117  2017   \n",
       "3082      BREST TRAIN STATION 0 days 00:46:04.645000         46.077417  2017   \n",
       "3080  ARMORIQUE REGIONAL PARK 5 days 01:33:43.371000       7293.722850  2017   \n",
       "3077  ARMORIQUE REGIONAL PARK 0 days 03:36:25.402000        216.423367  2017   \n",
       "3078  ARMORIQUE REGIONAL PARK 3 days 20:00:42.796000       5520.713267  2017   \n",
       "\n",
       "      Month        Date  nb_commas  \n",
       "3081      8  2017-08-15        3.0  \n",
       "3082      8  2017-08-15        2.0  \n",
       "3080      8  2017-08-10        2.0  \n",
       "3077      8  2017-08-05        2.0  \n",
       "3078      8  2017-08-05        2.0  "
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# FIRST CASE: FRENCH ADDRESSES (MOST OF MY DATA)\n",
    "\n",
    "# Step 1: For each Google address, I compute the number of \",\" \n",
    "    #nb_commas == 1 : We have only the zipcode, town name and country name\n",
    "    #nb_commas == 2 : The address is complete (i.e. we have the street name and number)\n",
    "place_visits_df[\"nb_commas\"] = place_visits_df[\"address_clean\"].str.count(\",\")\n",
    "place_visits_df.sort_values(\"longitude\").head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "89d55894",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>placeId</th>\n",
       "      <th>locationConfidence</th>\n",
       "      <th>startTimestamp</th>\n",
       "      <th>endTimestamp</th>\n",
       "      <th>placeVisitImportance</th>\n",
       "      <th>placeVisitType</th>\n",
       "      <th>name</th>\n",
       "      <th>address</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>...</th>\n",
       "      <th>name_clean</th>\n",
       "      <th>duration</th>\n",
       "      <th>duration_minutes</th>\n",
       "      <th>Year</th>\n",
       "      <th>Month</th>\n",
       "      <th>Date</th>\n",
       "      <th>nb_commas</th>\n",
       "      <th>Street</th>\n",
       "      <th>Zipcode &amp; town</th>\n",
       "      <th>Country</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3081</th>\n",
       "      <td>ChIJ4YbuA1y5FkgRnVJIe4VlqOk</td>\n",
       "      <td>89.39192</td>\n",
       "      <td>2017-08-15 13:54:46.243000+00:00</td>\n",
       "      <td>2017-08-15 15:06:45.530000+00:00</td>\n",
       "      <td>MAIN</td>\n",
       "      <td>SINGLE_PLACE</td>\n",
       "      <td>Square Alphonse Juin</td>\n",
       "      <td>10-, 14 Rue Frégate la Belle Poule, 29200 Bres...</td>\n",
       "      <td>48.386292</td>\n",
       "      <td>-4.483549</td>\n",
       "      <td>...</td>\n",
       "      <td>SQUARE ALPHONSE JUIN</td>\n",
       "      <td>0 days 01:11:59.287000</td>\n",
       "      <td>71.988117</td>\n",
       "      <td>2017</td>\n",
       "      <td>8</td>\n",
       "      <td>2017-08-15</td>\n",
       "      <td>3.0</td>\n",
       "      <td>10-</td>\n",
       "      <td>14 RUE FREGATE LA BELLE POULE</td>\n",
       "      <td>29200 BREST, FRANCE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3082</th>\n",
       "      <td>ChIJ0xnzm2G5FkgRSMgfQYsYEpM</td>\n",
       "      <td>99.66684</td>\n",
       "      <td>2017-08-15 15:09:49.427000+00:00</td>\n",
       "      <td>2017-08-15 15:55:54.072000+00:00</td>\n",
       "      <td>MAIN</td>\n",
       "      <td>SINGLE_PLACE</td>\n",
       "      <td>Brest Train Station</td>\n",
       "      <td>08 Place du 19ème Régiment d'Infanterie, 61134...</td>\n",
       "      <td>48.387659</td>\n",
       "      <td>-4.480156</td>\n",
       "      <td>...</td>\n",
       "      <td>BREST TRAIN STATION</td>\n",
       "      <td>0 days 00:46:04.645000</td>\n",
       "      <td>46.077417</td>\n",
       "      <td>2017</td>\n",
       "      <td>8</td>\n",
       "      <td>2017-08-15</td>\n",
       "      <td>2.0</td>\n",
       "      <td>08 PLACE DU 19EME REGIMENT D'INFANTERIE</td>\n",
       "      <td>61134-29211 BREST</td>\n",
       "      <td>FRANCE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3080</th>\n",
       "      <td>ChIJz2lWBU3IFkgRe0_6oTJ5Mpw</td>\n",
       "      <td>99.90851</td>\n",
       "      <td>2017-08-10 11:05:21.169000+00:00</td>\n",
       "      <td>2017-08-15 12:39:04.540000+00:00</td>\n",
       "      <td>MAIN</td>\n",
       "      <td>SINGLE_PLACE</td>\n",
       "      <td>Armorique Regional Park</td>\n",
       "      <td>15 Place aux Foires, 29590 Le Faou, France</td>\n",
       "      <td>48.293486</td>\n",
       "      <td>-4.178065</td>\n",
       "      <td>...</td>\n",
       "      <td>ARMORIQUE REGIONAL PARK</td>\n",
       "      <td>5 days 01:33:43.371000</td>\n",
       "      <td>7293.722850</td>\n",
       "      <td>2017</td>\n",
       "      <td>8</td>\n",
       "      <td>2017-08-10</td>\n",
       "      <td>2.0</td>\n",
       "      <td>15 PLACE AUX FOIRES</td>\n",
       "      <td>29590 LE FAOU</td>\n",
       "      <td>FRANCE</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                          placeId  locationConfidence  \\\n",
       "3081  ChIJ4YbuA1y5FkgRnVJIe4VlqOk            89.39192   \n",
       "3082  ChIJ0xnzm2G5FkgRSMgfQYsYEpM            99.66684   \n",
       "3080  ChIJz2lWBU3IFkgRe0_6oTJ5Mpw            99.90851   \n",
       "\n",
       "                       startTimestamp                     endTimestamp  \\\n",
       "3081 2017-08-15 13:54:46.243000+00:00 2017-08-15 15:06:45.530000+00:00   \n",
       "3082 2017-08-15 15:09:49.427000+00:00 2017-08-15 15:55:54.072000+00:00   \n",
       "3080 2017-08-10 11:05:21.169000+00:00 2017-08-15 12:39:04.540000+00:00   \n",
       "\n",
       "     placeVisitImportance placeVisitType                     name  \\\n",
       "3081                 MAIN   SINGLE_PLACE     Square Alphonse Juin   \n",
       "3082                 MAIN   SINGLE_PLACE      Brest Train Station   \n",
       "3080                 MAIN   SINGLE_PLACE  Armorique Regional Park   \n",
       "\n",
       "                                                address   latitude  longitude  \\\n",
       "3081  10-, 14 Rue Frégate la Belle Poule, 29200 Bres...  48.386292  -4.483549   \n",
       "3082  08 Place du 19ème Régiment d'Infanterie, 61134...  48.387659  -4.480156   \n",
       "3080         15 Place aux Foires, 29590 Le Faou, France  48.293486  -4.178065   \n",
       "\n",
       "      ...               name_clean               duration duration_minutes  \\\n",
       "3081  ...     SQUARE ALPHONSE JUIN 0 days 01:11:59.287000        71.988117   \n",
       "3082  ...      BREST TRAIN STATION 0 days 00:46:04.645000        46.077417   \n",
       "3080  ...  ARMORIQUE REGIONAL PARK 5 days 01:33:43.371000      7293.722850   \n",
       "\n",
       "      Year Month        Date  nb_commas  \\\n",
       "3081  2017     8  2017-08-15        3.0   \n",
       "3082  2017     8  2017-08-15        2.0   \n",
       "3080  2017     8  2017-08-10        2.0   \n",
       "\n",
       "                                       Street                  Zipcode & town  \\\n",
       "3081                                      10-   14 RUE FREGATE LA BELLE POULE   \n",
       "3082  08 PLACE DU 19EME REGIMENT D'INFANTERIE               61134-29211 BREST   \n",
       "3080                      15 PLACE AUX FOIRES                   29590 LE FAOU   \n",
       "\n",
       "                   Country  \n",
       "3081   29200 BREST, FRANCE  \n",
       "3082                FRANCE  \n",
       "3080                FRANCE  \n",
       "\n",
       "[3 rows x 23 columns]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Step 2: I split the French geographic information into different columns: Street (name and number), Zipcode & City name, Country\n",
    "new = place_visits_df[\"address_clean\"].str.split(\",\", n = 1, expand = True)\n",
    "place_visits_df[\"Street\"]= new[0]\n",
    "place_visits_df[\"Zipcode & town\"]= new[1]\n",
    "new2 = place_visits_df[\"Zipcode & town\"].str.split(\",\", n = 1, expand = True)\n",
    "place_visits_df[\"Zipcode & town\"]= new2[0]\n",
    "place_visits_df[\"Country\"]= new2[1]\n",
    "\n",
    "# It works quite for French addresses, except those that are incomplete (no street number and name)\n",
    "place_visits_df.sort_values(\"longitude\").head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "45660c2d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>placeId</th>\n",
       "      <th>locationConfidence</th>\n",
       "      <th>startTimestamp</th>\n",
       "      <th>endTimestamp</th>\n",
       "      <th>placeVisitImportance</th>\n",
       "      <th>placeVisitType</th>\n",
       "      <th>name</th>\n",
       "      <th>address</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>...</th>\n",
       "      <th>duration_minutes</th>\n",
       "      <th>Year</th>\n",
       "      <th>Month</th>\n",
       "      <th>Date</th>\n",
       "      <th>nb_commas</th>\n",
       "      <th>Street</th>\n",
       "      <th>Zipcode &amp; town</th>\n",
       "      <th>Country</th>\n",
       "      <th>Zipcode</th>\n",
       "      <th>City</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3081</th>\n",
       "      <td>ChIJ4YbuA1y5FkgRnVJIe4VlqOk</td>\n",
       "      <td>89.39192</td>\n",
       "      <td>2017-08-15 13:54:46.243000+00:00</td>\n",
       "      <td>2017-08-15 15:06:45.530000+00:00</td>\n",
       "      <td>MAIN</td>\n",
       "      <td>SINGLE_PLACE</td>\n",
       "      <td>Square Alphonse Juin</td>\n",
       "      <td>10-, 14 Rue Frégate la Belle Poule, 29200 Bres...</td>\n",
       "      <td>48.386292</td>\n",
       "      <td>-4.483549</td>\n",
       "      <td>...</td>\n",
       "      <td>71.988117</td>\n",
       "      <td>2017</td>\n",
       "      <td>8</td>\n",
       "      <td>2017-08-15</td>\n",
       "      <td>3.0</td>\n",
       "      <td>10-</td>\n",
       "      <td>14 RUE FREGATE LA BELLE POULE</td>\n",
       "      <td>29200 BREST, FRANCE</td>\n",
       "      <td>14 RU</td>\n",
       "      <td>E FREGATE LA BELLE POULE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3082</th>\n",
       "      <td>ChIJ0xnzm2G5FkgRSMgfQYsYEpM</td>\n",
       "      <td>99.66684</td>\n",
       "      <td>2017-08-15 15:09:49.427000+00:00</td>\n",
       "      <td>2017-08-15 15:55:54.072000+00:00</td>\n",
       "      <td>MAIN</td>\n",
       "      <td>SINGLE_PLACE</td>\n",
       "      <td>Brest Train Station</td>\n",
       "      <td>08 Place du 19ème Régiment d'Infanterie, 61134...</td>\n",
       "      <td>48.387659</td>\n",
       "      <td>-4.480156</td>\n",
       "      <td>...</td>\n",
       "      <td>46.077417</td>\n",
       "      <td>2017</td>\n",
       "      <td>8</td>\n",
       "      <td>2017-08-15</td>\n",
       "      <td>2.0</td>\n",
       "      <td>08 PLACE DU 19EME REGIMENT D'INFANTERIE</td>\n",
       "      <td>61134-29211 BREST</td>\n",
       "      <td>FRANCE</td>\n",
       "      <td>61134</td>\n",
       "      <td>-29211 BREST</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3080</th>\n",
       "      <td>ChIJz2lWBU3IFkgRe0_6oTJ5Mpw</td>\n",
       "      <td>99.90851</td>\n",
       "      <td>2017-08-10 11:05:21.169000+00:00</td>\n",
       "      <td>2017-08-15 12:39:04.540000+00:00</td>\n",
       "      <td>MAIN</td>\n",
       "      <td>SINGLE_PLACE</td>\n",
       "      <td>Armorique Regional Park</td>\n",
       "      <td>15 Place aux Foires, 29590 Le Faou, France</td>\n",
       "      <td>48.293486</td>\n",
       "      <td>-4.178065</td>\n",
       "      <td>...</td>\n",
       "      <td>7293.722850</td>\n",
       "      <td>2017</td>\n",
       "      <td>8</td>\n",
       "      <td>2017-08-10</td>\n",
       "      <td>2.0</td>\n",
       "      <td>15 PLACE AUX FOIRES</td>\n",
       "      <td>29590 LE FAOU</td>\n",
       "      <td>FRANCE</td>\n",
       "      <td>29590</td>\n",
       "      <td>LE FAOU</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3077</th>\n",
       "      <td>ChIJz2lWBU3IFkgRe0_6oTJ5Mpw</td>\n",
       "      <td>99.88017</td>\n",
       "      <td>2017-08-05 13:42:39.180000+00:00</td>\n",
       "      <td>2017-08-05 17:19:04.582000+00:00</td>\n",
       "      <td>MAIN</td>\n",
       "      <td>SINGLE_PLACE</td>\n",
       "      <td>Armorique Regional Park</td>\n",
       "      <td>15 Place aux Foires, 29590 Le Faou, France</td>\n",
       "      <td>48.293486</td>\n",
       "      <td>-4.178065</td>\n",
       "      <td>...</td>\n",
       "      <td>216.423367</td>\n",
       "      <td>2017</td>\n",
       "      <td>8</td>\n",
       "      <td>2017-08-05</td>\n",
       "      <td>2.0</td>\n",
       "      <td>15 PLACE AUX FOIRES</td>\n",
       "      <td>29590 LE FAOU</td>\n",
       "      <td>FRANCE</td>\n",
       "      <td>29590</td>\n",
       "      <td>LE FAOU</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3078</th>\n",
       "      <td>ChIJz2lWBU3IFkgRe0_6oTJ5Mpw</td>\n",
       "      <td>99.94573</td>\n",
       "      <td>2017-08-05 18:10:18.126000+00:00</td>\n",
       "      <td>2017-08-09 14:11:00.922000+00:00</td>\n",
       "      <td>MAIN</td>\n",
       "      <td>SINGLE_PLACE</td>\n",
       "      <td>Armorique Regional Park</td>\n",
       "      <td>15 Place aux Foires, 29590 Le Faou, France</td>\n",
       "      <td>48.293486</td>\n",
       "      <td>-4.178065</td>\n",
       "      <td>...</td>\n",
       "      <td>5520.713267</td>\n",
       "      <td>2017</td>\n",
       "      <td>8</td>\n",
       "      <td>2017-08-05</td>\n",
       "      <td>2.0</td>\n",
       "      <td>15 PLACE AUX FOIRES</td>\n",
       "      <td>29590 LE FAOU</td>\n",
       "      <td>FRANCE</td>\n",
       "      <td>29590</td>\n",
       "      <td>LE FAOU</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 25 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                          placeId  locationConfidence  \\\n",
       "3081  ChIJ4YbuA1y5FkgRnVJIe4VlqOk            89.39192   \n",
       "3082  ChIJ0xnzm2G5FkgRSMgfQYsYEpM            99.66684   \n",
       "3080  ChIJz2lWBU3IFkgRe0_6oTJ5Mpw            99.90851   \n",
       "3077  ChIJz2lWBU3IFkgRe0_6oTJ5Mpw            99.88017   \n",
       "3078  ChIJz2lWBU3IFkgRe0_6oTJ5Mpw            99.94573   \n",
       "\n",
       "                       startTimestamp                     endTimestamp  \\\n",
       "3081 2017-08-15 13:54:46.243000+00:00 2017-08-15 15:06:45.530000+00:00   \n",
       "3082 2017-08-15 15:09:49.427000+00:00 2017-08-15 15:55:54.072000+00:00   \n",
       "3080 2017-08-10 11:05:21.169000+00:00 2017-08-15 12:39:04.540000+00:00   \n",
       "3077 2017-08-05 13:42:39.180000+00:00 2017-08-05 17:19:04.582000+00:00   \n",
       "3078 2017-08-05 18:10:18.126000+00:00 2017-08-09 14:11:00.922000+00:00   \n",
       "\n",
       "     placeVisitImportance placeVisitType                     name  \\\n",
       "3081                 MAIN   SINGLE_PLACE     Square Alphonse Juin   \n",
       "3082                 MAIN   SINGLE_PLACE      Brest Train Station   \n",
       "3080                 MAIN   SINGLE_PLACE  Armorique Regional Park   \n",
       "3077                 MAIN   SINGLE_PLACE  Armorique Regional Park   \n",
       "3078                 MAIN   SINGLE_PLACE  Armorique Regional Park   \n",
       "\n",
       "                                                address   latitude  longitude  \\\n",
       "3081  10-, 14 Rue Frégate la Belle Poule, 29200 Bres...  48.386292  -4.483549   \n",
       "3082  08 Place du 19ème Régiment d'Infanterie, 61134...  48.387659  -4.480156   \n",
       "3080         15 Place aux Foires, 29590 Le Faou, France  48.293486  -4.178065   \n",
       "3077         15 Place aux Foires, 29590 Le Faou, France  48.293486  -4.178065   \n",
       "3078         15 Place aux Foires, 29590 Le Faou, France  48.293486  -4.178065   \n",
       "\n",
       "      ...  duration_minutes  Year Month        Date nb_commas  \\\n",
       "3081  ...         71.988117  2017     8  2017-08-15       3.0   \n",
       "3082  ...         46.077417  2017     8  2017-08-15       2.0   \n",
       "3080  ...       7293.722850  2017     8  2017-08-10       2.0   \n",
       "3077  ...        216.423367  2017     8  2017-08-05       2.0   \n",
       "3078  ...       5520.713267  2017     8  2017-08-05       2.0   \n",
       "\n",
       "                                       Street                  Zipcode & town  \\\n",
       "3081                                      10-   14 RUE FREGATE LA BELLE POULE   \n",
       "3082  08 PLACE DU 19EME REGIMENT D'INFANTERIE               61134-29211 BREST   \n",
       "3080                      15 PLACE AUX FOIRES                   29590 LE FAOU   \n",
       "3077                      15 PLACE AUX FOIRES                   29590 LE FAOU   \n",
       "3078                      15 PLACE AUX FOIRES                   29590 LE FAOU   \n",
       "\n",
       "                   Country Zipcode                      City  \n",
       "3081   29200 BREST, FRANCE   14 RU  E FREGATE LA BELLE POULE  \n",
       "3082                FRANCE   61134              -29211 BREST  \n",
       "3080                FRANCE   29590                   LE FAOU  \n",
       "3077                FRANCE   29590                   LE FAOU  \n",
       "3078                FRANCE   29590                   LE FAOU  \n",
       "\n",
       "[5 rows x 25 columns]"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# I make some adjustments for the French addresses that do not have any Google Maps street name and number\n",
    "place_visits_df[\"Country\"] = np.where((place_visits_df[\"address_clean\"].str.contains(\"FRANCE\")) & (place_visits_df[\"Country\"].isna()),\n",
    "                                      \"FRANCE\", place_visits_df[\"Country\"])\n",
    "\n",
    "place_visits_df[\"Zipcode & town\"] = np.where((place_visits_df[\"Country\"]==\"FRANCE\")  & (place_visits_df[\"nb_commas\"]==1),\n",
    "                                      place_visits_df[\"Street\"], place_visits_df[\"Zipcode & town\"])\n",
    "\n",
    "place_visits_df[\"Street\"] = np.where((place_visits_df[\"Country\"]==\"FRANCE\") & (place_visits_df[\"nb_commas\"]==1),\n",
    "                                     None, place_visits_df[\"Street\"])   \n",
    "\n",
    "place_visits_df[\"Zipcode\"]= place_visits_df[\"Zipcode & town\"].str[:6]\n",
    "place_visits_df[\"City\"]= place_visits_df[\"Zipcode & town\"].str[6:].str.lstrip()\n",
    "\n",
    "place_visits_df.sort_values(\"longitude\").head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "59da7a4e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>placeId</th>\n",
       "      <th>locationConfidence</th>\n",
       "      <th>startTimestamp</th>\n",
       "      <th>endTimestamp</th>\n",
       "      <th>placeVisitImportance</th>\n",
       "      <th>placeVisitType</th>\n",
       "      <th>name</th>\n",
       "      <th>address</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>...</th>\n",
       "      <th>duration_minutes</th>\n",
       "      <th>Year</th>\n",
       "      <th>Month</th>\n",
       "      <th>Date</th>\n",
       "      <th>nb_commas</th>\n",
       "      <th>Street</th>\n",
       "      <th>Zipcode &amp; town</th>\n",
       "      <th>Country</th>\n",
       "      <th>Zipcode</th>\n",
       "      <th>City</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4871</th>\n",
       "      <td>ChIJQ6MVplUZ2jERn1LmNH0DlDA</td>\n",
       "      <td>99.06076</td>\n",
       "      <td>2017-04-30 02:24:13.325000+00:00</td>\n",
       "      <td>2017-04-30 04:06:55.138000+00:00</td>\n",
       "      <td>MAIN</td>\n",
       "      <td>SINGLE_PLACE</td>\n",
       "      <td>Universal Studios Singapore</td>\n",
       "      <td>8 Sentosa Gateway, Singapore 098269</td>\n",
       "      <td>1.254042</td>\n",
       "      <td>103.823808</td>\n",
       "      <td>...</td>\n",
       "      <td>102.696883</td>\n",
       "      <td>2017</td>\n",
       "      <td>4</td>\n",
       "      <td>2017-04-30</td>\n",
       "      <td>1.0</td>\n",
       "      <td>8 SENTOSA GATEWAY</td>\n",
       "      <td>SINGAPORE 098269</td>\n",
       "      <td>None</td>\n",
       "      <td>SINGA</td>\n",
       "      <td>PORE 098269</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3882</th>\n",
       "      <td>ChIJK7xLl1gZ2jERP_GdUY9XNLo</td>\n",
       "      <td>92.29856</td>\n",
       "      <td>2017-01-14 14:12:01.632000+00:00</td>\n",
       "      <td>2017-01-14 14:23:13.180000+00:00</td>\n",
       "      <td>MAIN</td>\n",
       "      <td>SINGLE_PLACE</td>\n",
       "      <td>VivoCity</td>\n",
       "      <td>1 Harbourfront Walk, Singapore 098585</td>\n",
       "      <td>1.264241</td>\n",
       "      <td>103.822327</td>\n",
       "      <td>...</td>\n",
       "      <td>11.192467</td>\n",
       "      <td>2017</td>\n",
       "      <td>1</td>\n",
       "      <td>2017-01-14</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1 HARBOURFRONT WALK</td>\n",
       "      <td>SINGAPORE 098585</td>\n",
       "      <td>None</td>\n",
       "      <td>SINGA</td>\n",
       "      <td>PORE 098585</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4872</th>\n",
       "      <td>ChIJK7xLl1gZ2jERP_GdUY9XNLo</td>\n",
       "      <td>99.62569</td>\n",
       "      <td>2017-04-30 04:27:02.294000+00:00</td>\n",
       "      <td>2017-04-30 07:35:24.445000+00:00</td>\n",
       "      <td>MAIN</td>\n",
       "      <td>SINGLE_PLACE</td>\n",
       "      <td>VivoCity</td>\n",
       "      <td>1 Harbourfront Walk, Singapore 098585</td>\n",
       "      <td>1.264241</td>\n",
       "      <td>103.822327</td>\n",
       "      <td>...</td>\n",
       "      <td>188.369183</td>\n",
       "      <td>2017</td>\n",
       "      <td>4</td>\n",
       "      <td>2017-04-30</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1 HARBOURFRONT WALK</td>\n",
       "      <td>SINGAPORE 098585</td>\n",
       "      <td>None</td>\n",
       "      <td>SINGA</td>\n",
       "      <td>PORE 098585</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3 rows × 25 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                          placeId  locationConfidence  \\\n",
       "4871  ChIJQ6MVplUZ2jERn1LmNH0DlDA            99.06076   \n",
       "3882  ChIJK7xLl1gZ2jERP_GdUY9XNLo            92.29856   \n",
       "4872  ChIJK7xLl1gZ2jERP_GdUY9XNLo            99.62569   \n",
       "\n",
       "                       startTimestamp                     endTimestamp  \\\n",
       "4871 2017-04-30 02:24:13.325000+00:00 2017-04-30 04:06:55.138000+00:00   \n",
       "3882 2017-01-14 14:12:01.632000+00:00 2017-01-14 14:23:13.180000+00:00   \n",
       "4872 2017-04-30 04:27:02.294000+00:00 2017-04-30 07:35:24.445000+00:00   \n",
       "\n",
       "     placeVisitImportance placeVisitType                         name  \\\n",
       "4871                 MAIN   SINGLE_PLACE  Universal Studios Singapore   \n",
       "3882                 MAIN   SINGLE_PLACE                     VivoCity   \n",
       "4872                 MAIN   SINGLE_PLACE                     VivoCity   \n",
       "\n",
       "                                    address  latitude   longitude  ...  \\\n",
       "4871    8 Sentosa Gateway, Singapore 098269  1.254042  103.823808  ...   \n",
       "3882  1 Harbourfront Walk, Singapore 098585  1.264241  103.822327  ...   \n",
       "4872  1 Harbourfront Walk, Singapore 098585  1.264241  103.822327  ...   \n",
       "\n",
       "      duration_minutes  Year Month        Date nb_commas               Street  \\\n",
       "4871        102.696883  2017     4  2017-04-30       1.0    8 SENTOSA GATEWAY   \n",
       "3882         11.192467  2017     1  2017-01-14       1.0  1 HARBOURFRONT WALK   \n",
       "4872        188.369183  2017     4  2017-04-30       1.0  1 HARBOURFRONT WALK   \n",
       "\n",
       "         Zipcode & town  Country Zipcode         City  \n",
       "4871   SINGAPORE 098269     None   SINGA  PORE 098269  \n",
       "3882   SINGAPORE 098585     None   SINGA  PORE 098585  \n",
       "4872   SINGAPORE 098585     None   SINGA  PORE 098585  \n",
       "\n",
       "[3 rows x 25 columns]"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# SECOND CASE: SINGAPORE ADDRESSES. \n",
    "    # Many changes have to be made\n",
    "place_visits_df.sort_values(\"latitude\").head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "c1a3abdb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>placeId</th>\n",
       "      <th>locationConfidence</th>\n",
       "      <th>startTimestamp</th>\n",
       "      <th>endTimestamp</th>\n",
       "      <th>placeVisitImportance</th>\n",
       "      <th>placeVisitType</th>\n",
       "      <th>name</th>\n",
       "      <th>address</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>...</th>\n",
       "      <th>duration_minutes</th>\n",
       "      <th>Year</th>\n",
       "      <th>Month</th>\n",
       "      <th>Date</th>\n",
       "      <th>nb_commas</th>\n",
       "      <th>Street</th>\n",
       "      <th>Zipcode &amp; town</th>\n",
       "      <th>Country</th>\n",
       "      <th>Zipcode</th>\n",
       "      <th>City</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4871</th>\n",
       "      <td>ChIJQ6MVplUZ2jERn1LmNH0DlDA</td>\n",
       "      <td>99.06076</td>\n",
       "      <td>2017-04-30 02:24:13.325000+00:00</td>\n",
       "      <td>2017-04-30 04:06:55.138000+00:00</td>\n",
       "      <td>MAIN</td>\n",
       "      <td>SINGLE_PLACE</td>\n",
       "      <td>Universal Studios Singapore</td>\n",
       "      <td>8 Sentosa Gateway, Singapore 098269</td>\n",
       "      <td>1.254042</td>\n",
       "      <td>103.823808</td>\n",
       "      <td>...</td>\n",
       "      <td>102.696883</td>\n",
       "      <td>2017</td>\n",
       "      <td>4</td>\n",
       "      <td>2017-04-30</td>\n",
       "      <td>1.0</td>\n",
       "      <td>8 SENTOSA GATEWAY</td>\n",
       "      <td>SINGAPORE 098269</td>\n",
       "      <td>SINGAPORE</td>\n",
       "      <td>09826</td>\n",
       "      <td>SINGAPORE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3882</th>\n",
       "      <td>ChIJK7xLl1gZ2jERP_GdUY9XNLo</td>\n",
       "      <td>92.29856</td>\n",
       "      <td>2017-01-14 14:12:01.632000+00:00</td>\n",
       "      <td>2017-01-14 14:23:13.180000+00:00</td>\n",
       "      <td>MAIN</td>\n",
       "      <td>SINGLE_PLACE</td>\n",
       "      <td>VivoCity</td>\n",
       "      <td>1 Harbourfront Walk, Singapore 098585</td>\n",
       "      <td>1.264241</td>\n",
       "      <td>103.822327</td>\n",
       "      <td>...</td>\n",
       "      <td>11.192467</td>\n",
       "      <td>2017</td>\n",
       "      <td>1</td>\n",
       "      <td>2017-01-14</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1 HARBOURFRONT WALK</td>\n",
       "      <td>SINGAPORE 098585</td>\n",
       "      <td>SINGAPORE</td>\n",
       "      <td>09858</td>\n",
       "      <td>SINGAPORE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4872</th>\n",
       "      <td>ChIJK7xLl1gZ2jERP_GdUY9XNLo</td>\n",
       "      <td>99.62569</td>\n",
       "      <td>2017-04-30 04:27:02.294000+00:00</td>\n",
       "      <td>2017-04-30 07:35:24.445000+00:00</td>\n",
       "      <td>MAIN</td>\n",
       "      <td>SINGLE_PLACE</td>\n",
       "      <td>VivoCity</td>\n",
       "      <td>1 Harbourfront Walk, Singapore 098585</td>\n",
       "      <td>1.264241</td>\n",
       "      <td>103.822327</td>\n",
       "      <td>...</td>\n",
       "      <td>188.369183</td>\n",
       "      <td>2017</td>\n",
       "      <td>4</td>\n",
       "      <td>2017-04-30</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1 HARBOURFRONT WALK</td>\n",
       "      <td>SINGAPORE 098585</td>\n",
       "      <td>SINGAPORE</td>\n",
       "      <td>09858</td>\n",
       "      <td>SINGAPORE</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3 rows × 25 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                          placeId  locationConfidence  \\\n",
       "4871  ChIJQ6MVplUZ2jERn1LmNH0DlDA            99.06076   \n",
       "3882  ChIJK7xLl1gZ2jERP_GdUY9XNLo            92.29856   \n",
       "4872  ChIJK7xLl1gZ2jERP_GdUY9XNLo            99.62569   \n",
       "\n",
       "                       startTimestamp                     endTimestamp  \\\n",
       "4871 2017-04-30 02:24:13.325000+00:00 2017-04-30 04:06:55.138000+00:00   \n",
       "3882 2017-01-14 14:12:01.632000+00:00 2017-01-14 14:23:13.180000+00:00   \n",
       "4872 2017-04-30 04:27:02.294000+00:00 2017-04-30 07:35:24.445000+00:00   \n",
       "\n",
       "     placeVisitImportance placeVisitType                         name  \\\n",
       "4871                 MAIN   SINGLE_PLACE  Universal Studios Singapore   \n",
       "3882                 MAIN   SINGLE_PLACE                     VivoCity   \n",
       "4872                 MAIN   SINGLE_PLACE                     VivoCity   \n",
       "\n",
       "                                    address  latitude   longitude  ...  \\\n",
       "4871    8 Sentosa Gateway, Singapore 098269  1.254042  103.823808  ...   \n",
       "3882  1 Harbourfront Walk, Singapore 098585  1.264241  103.822327  ...   \n",
       "4872  1 Harbourfront Walk, Singapore 098585  1.264241  103.822327  ...   \n",
       "\n",
       "      duration_minutes  Year Month        Date nb_commas               Street  \\\n",
       "4871        102.696883  2017     4  2017-04-30       1.0    8 SENTOSA GATEWAY   \n",
       "3882         11.192467  2017     1  2017-01-14       1.0  1 HARBOURFRONT WALK   \n",
       "4872        188.369183  2017     4  2017-04-30       1.0  1 HARBOURFRONT WALK   \n",
       "\n",
       "         Zipcode & town    Country Zipcode       City  \n",
       "4871   SINGAPORE 098269  SINGAPORE   09826  SINGAPORE  \n",
       "3882   SINGAPORE 098585  SINGAPORE   09858  SINGAPORE  \n",
       "4872   SINGAPORE 098585  SINGAPORE   09858  SINGAPORE  \n",
       "\n",
       "[3 rows x 25 columns]"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Singapore focus\n",
    "place_visits_df[\"Country\"] = np.where(place_visits_df[\"address_clean\"].str.contains(\"SINGAPORE\"), \n",
    "                                      \"SINGAPORE\", place_visits_df[\"Country\"])\n",
    "\n",
    "place_visits_df[\"City\"] = np.where(place_visits_df[\"Country\"]==\"SINGAPORE\", \"SINGAPORE\", place_visits_df[\"City\"])\n",
    "\n",
    "place_visits_df[\"Zipcode\"] = np.where(place_visits_df[\"Country\"]==\"SINGAPORE\", place_visits_df[\"Zipcode & town\"].str[-7:-1], place_visits_df[\"Zipcode\"])\n",
    "\n",
    "place_visits_df.sort_values(\"latitude\").head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "3f79e651",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>placeId</th>\n",
       "      <th>locationConfidence</th>\n",
       "      <th>startTimestamp</th>\n",
       "      <th>endTimestamp</th>\n",
       "      <th>placeVisitImportance</th>\n",
       "      <th>placeVisitType</th>\n",
       "      <th>name</th>\n",
       "      <th>address</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>...</th>\n",
       "      <th>duration_minutes</th>\n",
       "      <th>Year</th>\n",
       "      <th>Month</th>\n",
       "      <th>Date</th>\n",
       "      <th>nb_commas</th>\n",
       "      <th>Street</th>\n",
       "      <th>Zipcode &amp; town</th>\n",
       "      <th>Country</th>\n",
       "      <th>Zipcode</th>\n",
       "      <th>City</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>5765</th>\n",
       "      <td>ChIJ34BEgoRTFTURm9sPe3FbyOs</td>\n",
       "      <td>75.97877</td>\n",
       "      <td>2018-10-05 05:55:46.176000+00:00</td>\n",
       "      <td>2018-10-05 07:27:06.242000+00:00</td>\n",
       "      <td>MAIN</td>\n",
       "      <td>SINGLE_PLACE</td>\n",
       "      <td>Glover Garden</td>\n",
       "      <td>日本、〒850-0931 長崎県長崎市南山手町８−１</td>\n",
       "      <td>32.734080</td>\n",
       "      <td>129.869521</td>\n",
       "      <td>...</td>\n",
       "      <td>91.334433</td>\n",
       "      <td>2018</td>\n",
       "      <td>10</td>\n",
       "      <td>2018-10-05</td>\n",
       "      <td>0.0</td>\n",
       "      <td>850-0931 81</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5766</th>\n",
       "      <td>ChIJQy20Fi5TFTURj3L4pJUnG_4</td>\n",
       "      <td>82.28539</td>\n",
       "      <td>2018-10-05 08:03:14.238000+00:00</td>\n",
       "      <td>2018-10-05 08:10:46.702000+00:00</td>\n",
       "      <td>MAIN</td>\n",
       "      <td>SINGLE_PLACE</td>\n",
       "      <td>Dutch Slope</td>\n",
       "      <td>日本、〒850-0911 長崎県長崎市東山手町</td>\n",
       "      <td>32.738323</td>\n",
       "      <td>129.872951</td>\n",
       "      <td>...</td>\n",
       "      <td>7.541067</td>\n",
       "      <td>2018</td>\n",
       "      <td>10</td>\n",
       "      <td>2018-10-05</td>\n",
       "      <td>0.0</td>\n",
       "      <td>850-0911</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5767</th>\n",
       "      <td>ChIJSfofOkdTFTURXov-p4VbjtQ</td>\n",
       "      <td>72.92718</td>\n",
       "      <td>2018-10-05 08:47:24.684000+00:00</td>\n",
       "      <td>2018-10-05 09:08:40.059000+00:00</td>\n",
       "      <td>MAIN</td>\n",
       "      <td>SINGLE_PLACE</td>\n",
       "      <td>Megane Bridge</td>\n",
       "      <td>日本、〒850-0874 長崎県長崎市魚の町</td>\n",
       "      <td>32.747146</td>\n",
       "      <td>129.880102</td>\n",
       "      <td>...</td>\n",
       "      <td>21.256250</td>\n",
       "      <td>2018</td>\n",
       "      <td>10</td>\n",
       "      <td>2018-10-05</td>\n",
       "      <td>0.0</td>\n",
       "      <td>850-0874</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3 rows × 25 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                          placeId  locationConfidence  \\\n",
       "5765  ChIJ34BEgoRTFTURm9sPe3FbyOs            75.97877   \n",
       "5766  ChIJQy20Fi5TFTURj3L4pJUnG_4            82.28539   \n",
       "5767  ChIJSfofOkdTFTURXov-p4VbjtQ            72.92718   \n",
       "\n",
       "                       startTimestamp                     endTimestamp  \\\n",
       "5765 2018-10-05 05:55:46.176000+00:00 2018-10-05 07:27:06.242000+00:00   \n",
       "5766 2018-10-05 08:03:14.238000+00:00 2018-10-05 08:10:46.702000+00:00   \n",
       "5767 2018-10-05 08:47:24.684000+00:00 2018-10-05 09:08:40.059000+00:00   \n",
       "\n",
       "     placeVisitImportance placeVisitType           name  \\\n",
       "5765                 MAIN   SINGLE_PLACE  Glover Garden   \n",
       "5766                 MAIN   SINGLE_PLACE    Dutch Slope   \n",
       "5767                 MAIN   SINGLE_PLACE  Megane Bridge   \n",
       "\n",
       "                         address   latitude   longitude  ...  \\\n",
       "5765  日本、〒850-0931 長崎県長崎市南山手町８−１  32.734080  129.869521  ...   \n",
       "5766     日本、〒850-0911 長崎県長崎市東山手町  32.738323  129.872951  ...   \n",
       "5767      日本、〒850-0874 長崎県長崎市魚の町  32.747146  129.880102  ...   \n",
       "\n",
       "      duration_minutes  Year Month        Date nb_commas       Street  \\\n",
       "5765         91.334433  2018    10  2018-10-05       0.0  850-0931 81   \n",
       "5766          7.541067  2018    10  2018-10-05       0.0    850-0911    \n",
       "5767         21.256250  2018    10  2018-10-05       0.0    850-0874    \n",
       "\n",
       "      Zipcode & town  Country Zipcode  City  \n",
       "5765            None     None    None  None  \n",
       "5766            None     None    None  None  \n",
       "5767            None     None    None  None  \n",
       "\n",
       "[3 rows x 25 columns]"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# THIRD CASE: JAPAN\n",
    "    # Unfortunatelty Python strings do not recognize kanjis so I cannot do much about it.\n",
    "place_visits_df[place_visits_df[\"Year\"]==2018].sort_values(\"latitude\").head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "dc1c3b05",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>placeId</th>\n",
       "      <th>locationConfidence</th>\n",
       "      <th>startTimestamp</th>\n",
       "      <th>endTimestamp</th>\n",
       "      <th>placeVisitImportance</th>\n",
       "      <th>placeVisitType</th>\n",
       "      <th>name</th>\n",
       "      <th>address</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>...</th>\n",
       "      <th>duration_minutes</th>\n",
       "      <th>Year</th>\n",
       "      <th>Month</th>\n",
       "      <th>Date</th>\n",
       "      <th>nb_commas</th>\n",
       "      <th>Street</th>\n",
       "      <th>Zipcode &amp; town</th>\n",
       "      <th>Country</th>\n",
       "      <th>Zipcode</th>\n",
       "      <th>City</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>5765</th>\n",
       "      <td>ChIJ34BEgoRTFTURm9sPe3FbyOs</td>\n",
       "      <td>75.97877</td>\n",
       "      <td>2018-10-05 05:55:46.176000+00:00</td>\n",
       "      <td>2018-10-05 07:27:06.242000+00:00</td>\n",
       "      <td>MAIN</td>\n",
       "      <td>SINGLE_PLACE</td>\n",
       "      <td>Glover Garden</td>\n",
       "      <td>日本、〒850-0931 長崎県長崎市南山手町８−１</td>\n",
       "      <td>32.734080</td>\n",
       "      <td>129.869521</td>\n",
       "      <td>...</td>\n",
       "      <td>91.334433</td>\n",
       "      <td>2018</td>\n",
       "      <td>10</td>\n",
       "      <td>2018-10-05</td>\n",
       "      <td>0.0</td>\n",
       "      <td>850-0931 81</td>\n",
       "      <td>None</td>\n",
       "      <td>JAPAN</td>\n",
       "      <td>850-0931</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5766</th>\n",
       "      <td>ChIJQy20Fi5TFTURj3L4pJUnG_4</td>\n",
       "      <td>82.28539</td>\n",
       "      <td>2018-10-05 08:03:14.238000+00:00</td>\n",
       "      <td>2018-10-05 08:10:46.702000+00:00</td>\n",
       "      <td>MAIN</td>\n",
       "      <td>SINGLE_PLACE</td>\n",
       "      <td>Dutch Slope</td>\n",
       "      <td>日本、〒850-0911 長崎県長崎市東山手町</td>\n",
       "      <td>32.738323</td>\n",
       "      <td>129.872951</td>\n",
       "      <td>...</td>\n",
       "      <td>7.541067</td>\n",
       "      <td>2018</td>\n",
       "      <td>10</td>\n",
       "      <td>2018-10-05</td>\n",
       "      <td>0.0</td>\n",
       "      <td>850-0911</td>\n",
       "      <td>None</td>\n",
       "      <td>JAPAN</td>\n",
       "      <td>850-0911</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5767</th>\n",
       "      <td>ChIJSfofOkdTFTURXov-p4VbjtQ</td>\n",
       "      <td>72.92718</td>\n",
       "      <td>2018-10-05 08:47:24.684000+00:00</td>\n",
       "      <td>2018-10-05 09:08:40.059000+00:00</td>\n",
       "      <td>MAIN</td>\n",
       "      <td>SINGLE_PLACE</td>\n",
       "      <td>Megane Bridge</td>\n",
       "      <td>日本、〒850-0874 長崎県長崎市魚の町</td>\n",
       "      <td>32.747146</td>\n",
       "      <td>129.880102</td>\n",
       "      <td>...</td>\n",
       "      <td>21.256250</td>\n",
       "      <td>2018</td>\n",
       "      <td>10</td>\n",
       "      <td>2018-10-05</td>\n",
       "      <td>0.0</td>\n",
       "      <td>850-0874</td>\n",
       "      <td>None</td>\n",
       "      <td>JAPAN</td>\n",
       "      <td>850-0874</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3 rows × 25 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                          placeId  locationConfidence  \\\n",
       "5765  ChIJ34BEgoRTFTURm9sPe3FbyOs            75.97877   \n",
       "5766  ChIJQy20Fi5TFTURj3L4pJUnG_4            82.28539   \n",
       "5767  ChIJSfofOkdTFTURXov-p4VbjtQ            72.92718   \n",
       "\n",
       "                       startTimestamp                     endTimestamp  \\\n",
       "5765 2018-10-05 05:55:46.176000+00:00 2018-10-05 07:27:06.242000+00:00   \n",
       "5766 2018-10-05 08:03:14.238000+00:00 2018-10-05 08:10:46.702000+00:00   \n",
       "5767 2018-10-05 08:47:24.684000+00:00 2018-10-05 09:08:40.059000+00:00   \n",
       "\n",
       "     placeVisitImportance placeVisitType           name  \\\n",
       "5765                 MAIN   SINGLE_PLACE  Glover Garden   \n",
       "5766                 MAIN   SINGLE_PLACE    Dutch Slope   \n",
       "5767                 MAIN   SINGLE_PLACE  Megane Bridge   \n",
       "\n",
       "                         address   latitude   longitude  ...  \\\n",
       "5765  日本、〒850-0931 長崎県長崎市南山手町８−１  32.734080  129.869521  ...   \n",
       "5766     日本、〒850-0911 長崎県長崎市東山手町  32.738323  129.872951  ...   \n",
       "5767      日本、〒850-0874 長崎県長崎市魚の町  32.747146  129.880102  ...   \n",
       "\n",
       "      duration_minutes  Year Month        Date nb_commas       Street  \\\n",
       "5765         91.334433  2018    10  2018-10-05       0.0  850-0931 81   \n",
       "5766          7.541067  2018    10  2018-10-05       0.0    850-0911    \n",
       "5767         21.256250  2018    10  2018-10-05       0.0    850-0874    \n",
       "\n",
       "      Zipcode & town  Country   Zipcode  City  \n",
       "5765            None    JAPAN  850-0931  None  \n",
       "5766            None    JAPAN  850-0911  None  \n",
       "5767            None    JAPAN  850-0874  None  \n",
       "\n",
       "[3 rows x 25 columns]"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Improve addresses for Japan\n",
    "''' This code does not work in Python (kanji not recognized)\n",
    "place_visits_df[\"Country\"] = np.where(place_visits_df[\"address_clean\"].str.contains(\"日本\"), \n",
    "                                      \"JAPAN\", place_visits_df[\"Country\"])\n",
    "'''\n",
    "\n",
    "place_visits_df[\"Country\"] = np.where((place_visits_df[\"Date\"]>= datetime.date(2018,7,2)) & (place_visits_df[\"Date\"]<= datetime.date(2018,12,30)), \n",
    "                                      \"JAPAN\", place_visits_df[\"Country\"])\n",
    "\n",
    "place_visits_df[\"Zipcode\"] = np.where(place_visits_df[\"Country\"]==\"JAPAN\", place_visits_df[\"address_clean\"].str[:8], place_visits_df[\"Zipcode\"])\n",
    "\n",
    "\n",
    "place_visits_df[place_visits_df[\"Year\"]==2018].sort_values(\"latitude\").head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "c33c545b",
   "metadata": {},
   "outputs": [],
   "source": [
    "# OTHER COUNTRIES\n",
    "place_visits_df[\"Country\"] = np.where(place_visits_df[\"address_clean\"].str.contains(\"SPAIN\"), \n",
    "                                      \"SPAIN\", place_visits_df[\"Country\"])\n",
    "\n",
    "place_visits_df[\"Country\"] = np.where(place_visits_df[\"address_clean\"].str.contains(\"NEDERLAND\"), \n",
    "                                      \"NETHERLANDS\", place_visits_df[\"Country\"])\n",
    "\n",
    "place_visits_df[\"Country\"] = np.where((place_visits_df[\"address_clean\"].str.contains(\"BELGIE\")) | (place_visits_df[\"address_clean\"].str.contains(\"BELGIQUE\")), \n",
    "                                      \"BELGIUM\", place_visits_df[\"Country\"])\n",
    "                                      \n",
    "place_visits_df[\"Country\"] = np.where(place_visits_df[\"address_clean\"].str.contains(\"SUISSE\"), \n",
    "                                      \"SWITZERLAND\", place_visits_df[\"Country\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "a065e263",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Remove the last columns that are of no use\n",
    "place_visits_df = place_visits_df.drop([\"name\", \"address\"], axis = 1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d1883bab",
   "metadata": {},
   "source": [
    "<div style=\"background-color: RGB(0,150,250);\" >\n",
    "<h2 style=\"margin: auto; padding: 20px; color:#fff; \">6. Export the dataframes to Excel (csv format)</h2> <a class=\"anchor\" id=\"6\"></a>\n",
    "</div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "594264e3",
   "metadata": {},
   "outputs": [],
   "source": [
    "place_visits_df.to_csv(\"place_visits.csv\", index=False, sep = \";\")\n",
    "activities_df.to_csv(\"activities.csv\", index = False, sep = \";\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e5351984",
   "metadata": {},
   "source": [
    "<div style=\"background-color: RGB(0,150,250);\" >\n",
    "<h2 style=\"margin: auto; padding: 20px; color:#fff; \">7. Insights from the data</h2> <a class=\"anchor\" id=\"7\"></a>\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3b58a54e",
   "metadata": {},
   "source": [
    "I want to answer 2 main questions:\n",
    "- How many times did I do swimming and dancing workouts in the past? \n",
    "- What is the average duration of my workouts?"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cc3676da",
   "metadata": {},
   "source": [
    "### 1) Number of times I did swimming and dancing workouts <a class=\"anchor\" id=\"7.1\"></a>\n",
    "\n",
    "In this section, I derive my swimming and dancing workouts from Google Maps data thanks to the name of the places I visited.\n",
    "- Swimming workouts: The Google name of the place I visited contains the word \"piscine\" (swimming pool in French)\n",
    "- Dance workouts: The Google name of the place is in the list of the dance studios I attended in the past"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "29db2b05",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create a \"Sports\" variable that takes the value:\n",
    "    # \"swimming\" if the place has \"piscine\" (swimming pool in France) in its name\n",
    "    # \"dance\" if the name of the place contains the name of the studio dances I attended\n",
    "    \n",
    "place_visits_df[\"Sports\"] = np.where(place_visits_df[\"name_clean\"].str.contains(\"PISCINE\", na = False),\n",
    "                                    \"SWIMMING\", None)\n",
    "\n",
    "place_visits_df[\"Sports\"] = np.where(place_visits_df[\"name_clean\"].isin([\"STUDIO HARMONIC\", \n",
    "                                                                        \"ZHENGHUA COMMUNITY CLUB\",\n",
    "                                                                       \"BISHAN COMMUNITY CLUB\",\n",
    "                                                                       \"313@SOMERSET\",\n",
    "                                                                       \"ORCHARD CENTRAL\",\n",
    "                                                                       \"DANCE ON US\",\n",
    "                                                                       \"DANCE FACTORY SINGAPORE\"]),\n",
    "                                    \"DANCE\", place_visits_df[\"Sports\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "12487408",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Year</th>\n",
       "      <th>Sports</th>\n",
       "      <th>value</th>\n",
       "      <th>value_month</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2017</td>\n",
       "      <td>DANCE</td>\n",
       "      <td>19</td>\n",
       "      <td>1.583333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2018</td>\n",
       "      <td>DANCE</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2019</td>\n",
       "      <td>DANCE</td>\n",
       "      <td>9</td>\n",
       "      <td>0.750000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2020</td>\n",
       "      <td>DANCE</td>\n",
       "      <td>10</td>\n",
       "      <td>0.833333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2021</td>\n",
       "      <td>DANCE</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2022</td>\n",
       "      <td>DANCE</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2023</td>\n",
       "      <td>DANCE</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>2017</td>\n",
       "      <td>SWIMMING</td>\n",
       "      <td>10</td>\n",
       "      <td>0.833333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>2018</td>\n",
       "      <td>SWIMMING</td>\n",
       "      <td>2</td>\n",
       "      <td>0.166667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>2019</td>\n",
       "      <td>SWIMMING</td>\n",
       "      <td>2</td>\n",
       "      <td>0.166667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>2020</td>\n",
       "      <td>SWIMMING</td>\n",
       "      <td>13</td>\n",
       "      <td>1.083333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>2021</td>\n",
       "      <td>SWIMMING</td>\n",
       "      <td>40</td>\n",
       "      <td>3.333333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>2022</td>\n",
       "      <td>SWIMMING</td>\n",
       "      <td>60</td>\n",
       "      <td>5.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>2023</td>\n",
       "      <td>SWIMMING</td>\n",
       "      <td>30</td>\n",
       "      <td>2.500000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    Year    Sports  value  value_month\n",
       "0   2017     DANCE     19     1.583333\n",
       "1   2018     DANCE      0     0.000000\n",
       "2   2019     DANCE      9     0.750000\n",
       "3   2020     DANCE     10     0.833333\n",
       "4   2021     DANCE      0     0.000000\n",
       "5   2022     DANCE      0     0.000000\n",
       "6   2023     DANCE      0     0.000000\n",
       "7   2017  SWIMMING     10     0.833333\n",
       "8   2018  SWIMMING      2     0.166667\n",
       "9   2019  SWIMMING      2     0.166667\n",
       "10  2020  SWIMMING     13     1.083333\n",
       "11  2021  SWIMMING     40     3.333333\n",
       "12  2022  SWIMMING     60     5.000000\n",
       "13  2023  SWIMMING     30     2.500000"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# YEAR ANALYSIS: For every year, I count the number of times I went to the swimming pool and dance classes\n",
    "sports_year = place_visits_df.groupby(['Year', 'Sports']).size().unstack(fill_value=0)\n",
    "sports_year.rename(columns={'DANCE': 'sport_DANCE', 'SWIMMING': 'sport_SWIMMING'}, inplace=True)\n",
    "sports_year.reset_index(inplace=True)\n",
    "\n",
    "sports_year = pd.wide_to_long(sports_year,\n",
    "               stubnames=['sport'],\n",
    "               i='Year',\n",
    "               j='Sports',\n",
    "               sep = '_',\n",
    "               suffix='.+')\n",
    "\n",
    "sports_year.reset_index(inplace=True)\n",
    "sports_year.rename(columns={'sport': 'value'}, inplace=True)\n",
    "sports_year[\"value_month\"] = sports_year[\"value\"] /12 # to get a monthly average\n",
    "sports_year"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "c339f7a7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Text(0.5, 1.0, 'Number of occurences per year')]"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Lineplot : Year\n",
    "sns.set_style(\"dark\")\n",
    "palette = sns.color_palette(\"mako_r\", 2)\n",
    "sns.lineplot(data = sports_year, x= \"Year\", y= \"value\", hue = \"Sports\", ci= None,\n",
    "             markers=True, style=\"Sports\", palette = palette).set(title='Number of occurences per year')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ed487476",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-block alert-warning\">\n",
    "<b>Learnings:</b> \n",
    "<b>My swimming and dancing activities seem to follow the same pattern from 2017 to 2019, and then start diverging drastically.</b> Indeed, from 2017 to 2019, I was still a student so if I had time for one activity, I also had some time for the other. On the other hand, if I was too busy and working abroad, I generally had no time for neither swimming nor dancing (cf. low points in 2018).\n",
    "    \n",
    "When the Covid19 crisis broke out early 2020, I completely stopped taking dance classes. Indeed, the dance studios were shut down during a very long period, and I never went back (cf. 2021 and 2022 flat dashed green lines). On the contrary, I increased dramatically the number of my swimming workouts (x6 increase between 2020 and 2022). In 2022, I went 60 times, i.e. at least once a week.\n",
    "</div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "eccd5d3c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Year</th>\n",
       "      <th>Month</th>\n",
       "      <th>Sports</th>\n",
       "      <th>value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2017</td>\n",
       "      <td>1</td>\n",
       "      <td>DANCE</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2017</td>\n",
       "      <td>1</td>\n",
       "      <td>SWIMMING</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2017</td>\n",
       "      <td>2</td>\n",
       "      <td>DANCE</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2017</td>\n",
       "      <td>2</td>\n",
       "      <td>SWIMMING</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2017</td>\n",
       "      <td>3</td>\n",
       "      <td>DANCE</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75</th>\n",
       "      <td>2023</td>\n",
       "      <td>6</td>\n",
       "      <td>SWIMMING</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>76</th>\n",
       "      <td>2023</td>\n",
       "      <td>7</td>\n",
       "      <td>DANCE</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>77</th>\n",
       "      <td>2023</td>\n",
       "      <td>7</td>\n",
       "      <td>SWIMMING</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>78</th>\n",
       "      <td>2023</td>\n",
       "      <td>8</td>\n",
       "      <td>DANCE</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>79</th>\n",
       "      <td>2023</td>\n",
       "      <td>8</td>\n",
       "      <td>SWIMMING</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>80 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    Year  Month    Sports  value\n",
       "0   2017      1     DANCE      2\n",
       "1   2017      1  SWIMMING      0\n",
       "2   2017      2     DANCE      4\n",
       "3   2017      2  SWIMMING      0\n",
       "4   2017      3     DANCE      2\n",
       "..   ...    ...       ...    ...\n",
       "75  2023      6  SWIMMING      8\n",
       "76  2023      7     DANCE      0\n",
       "77  2023      7  SWIMMING      5\n",
       "78  2023      8     DANCE      0\n",
       "79  2023      8  SWIMMING      2\n",
       "\n",
       "[80 rows x 4 columns]"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#MONTH ANALYSIS: For every month and every year, I count the number of times I went to the swimming pool and dance classes\n",
    "sports_month = place_visits_df.groupby(['Year', 'Month', 'Sports']).size().unstack(fill_value=0)\n",
    "sports_month.rename(columns={'DANCE': 'sport_DANCE', 'SWIMMING': 'sport_SWIMMING'}, inplace=True)\n",
    "sports_month.reset_index(inplace=True)\n",
    "\n",
    "sports_month = pd.wide_to_long(sports_month,\n",
    "               stubnames=['sport'],\n",
    "               i=['Year', 'Month'],\n",
    "               j='Sports',\n",
    "               sep = '_',\n",
    "               suffix='.+')\n",
    "\n",
    "sports_month.reset_index(inplace=True)\n",
    "sports_month.rename(columns={'sport': 'value'}, inplace=True)\n",
    "sports_month"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "d89fa646",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x11500a8ccd0>"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1081x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Lineplot (monthly analysis)\n",
    "sns.set_style(\"dark\")\n",
    "palette_7 = sns.color_palette(\"magma\", 7)\n",
    "sns.relplot(data = sports_month, x= \"Month\", y= \"value\", hue = \"Year\", ci= None,\n",
    "            kind= \"line\", col=\"Sports\",\n",
    "             markers=True, style=\"Year\", palette = palette_7)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "30f3080a",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-block alert-warning\">\n",
    "<b>Learnings:</b> \n",
    "    <b>My dancing lessons have been very erratic.</b> I have been quite consistent during the first semester of 2017 (dark purple line), \n",
    "when I was taking KPop dance classes during my exchange in Singapore (2 to 3 times a month). Then I stopped and I tried to resume dance classes after finishing my studies at the end of 2019 (classes twice a week in November). But then I gave up due to 2 external events: i) the transportation strikes that took place in France in December 2019 and January 2020 that prevented me going to the dance studio, ii) the first Covid19 lockdown in earch March 2020.\n",
    "\n",
    "<b>Regarding swimming, some seasonality seems to be at play.</b> I almost never go swimming during winter, and I most active in summer when there is no lockdown. Moreover, during summer, my nearby swimming pools close at 10PM (instead of 7PM) so it is more convenient to go after work. In August 2022, I reached my record: I went 18 times, i.e. every 1.5 days !\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fac3b1d7",
   "metadata": {},
   "source": [
    "### 2) Workout durations <a class=\"anchor\" id=\"7.2\"></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "630859b3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Sports    Year\n",
       "DANCE     2017    117.253164\n",
       "          2019    106.997198\n",
       "          2020    106.295742\n",
       "SWIMMING  2017     70.542805\n",
       "          2018     63.563200\n",
       "          2019     52.293600\n",
       "          2020     71.194932\n",
       "          2021     76.415544\n",
       "          2022     78.297432\n",
       "          2023     74.588189\n",
       "Name: duration_minutes, dtype: float64"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Compute the average time (in minutes) spent at the swimming pool / dance studio\n",
    "place_visits_df.groupby([\"Sports\", \"Year\"])[\"duration_minutes\"].mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "368db7b2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Text(0.5, 1.0, 'Duration in minutes')]"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Boxplot: Distribution time spent for each activity over time\n",
    "sns.boxplot(data = place_visits_df, x=\"Sports\", y= \"duration_minutes\", hue = \"Year\", palette = \"rocket\").set(title='Duration in minutes')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9173ab22",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-block alert-warning\">\n",
    "<b>Learnings:</b> \n",
    "<b>The duration of my workouts was pretty stable over time</b>. On average, I spent 1 hour and a half at the dance studio.\n",
    "This makes sense since my dance classes usually last 90 minutes. \n",
    "Regarding swimming, the duration of my workouts has slightly increased (52 minutes to 78 minutes on average between 2018 and 2022).\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "30e71806",
   "metadata": {},
   "source": [
    "### 3) Places visited during my Samsung Health workout  <a class=\"anchor\" id=\"7.3\"></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "df12f20d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Date</th>\n",
       "      <th>name_clean</th>\n",
       "      <th>address_clean</th>\n",
       "      <th>Zipcode</th>\n",
       "      <th>City</th>\n",
       "      <th>centerLat</th>\n",
       "      <th>centerLng</th>\n",
       "      <th>Zipcode &amp; town</th>\n",
       "      <th>Country</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>5470</th>\n",
       "      <td>2022-09-16</td>\n",
       "      <td>HOTEL TARONGETA</td>\n",
       "      <td>PARATGE TARONGETA, S/N, 17488 CADAQUES, GIRONA...</td>\n",
       "      <td>S/N</td>\n",
       "      <td></td>\n",
       "      <td>42.290032</td>\n",
       "      <td>3.273662</td>\n",
       "      <td>S/N</td>\n",
       "      <td>SPAIN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5461</th>\n",
       "      <td>2022-09-15</td>\n",
       "      <td>HOTEL TARONGETA</td>\n",
       "      <td>PARATGE TARONGETA, S/N, 17488 CADAQUES, GIRONA...</td>\n",
       "      <td>S/N</td>\n",
       "      <td></td>\n",
       "      <td>42.290061</td>\n",
       "      <td>3.273581</td>\n",
       "      <td>S/N</td>\n",
       "      <td>SPAIN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5453</th>\n",
       "      <td>2022-09-14</td>\n",
       "      <td>HOTEL GRI-MAR</td>\n",
       "      <td>N-260, 165, BAJO, 17490 LA VILA, GIRONA, SPAIN</td>\n",
       "      <td>165</td>\n",
       "      <td></td>\n",
       "      <td>42.370081</td>\n",
       "      <td>3.153537</td>\n",
       "      <td>165</td>\n",
       "      <td>SPAIN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5443</th>\n",
       "      <td>2022-09-13</td>\n",
       "      <td>HOTEL GRI-MAR</td>\n",
       "      <td>N-260, 165, BAJO, 17490 LA VILA, GIRONA, SPAIN</td>\n",
       "      <td>165</td>\n",
       "      <td></td>\n",
       "      <td>42.370103</td>\n",
       "      <td>3.153513</td>\n",
       "      <td>165</td>\n",
       "      <td>SPAIN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>766</th>\n",
       "      <td>2017-07-02</td>\n",
       "      <td>IBIS STYLES SINGAPORE ON MACPHERSON</td>\n",
       "      <td>401 MACPHERSON RD, SINGAPORE 368215</td>\n",
       "      <td>36821</td>\n",
       "      <td>SINGAPORE</td>\n",
       "      <td>1.331761</td>\n",
       "      <td>103.879298</td>\n",
       "      <td>SINGAPORE 368215</td>\n",
       "      <td>SINGAPORE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>759</th>\n",
       "      <td>2017-06-30</td>\n",
       "      <td>IBIS STYLES SINGAPORE ON MACPHERSON</td>\n",
       "      <td>401 MACPHERSON RD, SINGAPORE 368215</td>\n",
       "      <td>36821</td>\n",
       "      <td>SINGAPORE</td>\n",
       "      <td>1.331767</td>\n",
       "      <td>103.879257</td>\n",
       "      <td>SINGAPORE 368215</td>\n",
       "      <td>SINGAPORE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>753</th>\n",
       "      <td>2017-06-29</td>\n",
       "      <td>None</td>\n",
       "      <td>233 PAYA LEBAR RD, SINGAPORE 409044</td>\n",
       "      <td>40904</td>\n",
       "      <td>SINGAPORE</td>\n",
       "      <td>1.329978</td>\n",
       "      <td>103.888539</td>\n",
       "      <td>SINGAPORE 409044</td>\n",
       "      <td>SINGAPORE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>751</th>\n",
       "      <td>2017-06-27</td>\n",
       "      <td>LE CRESCENDO</td>\n",
       "      <td>233 PAYA LEBAR RD, SINGAPORE 409044</td>\n",
       "      <td>40904</td>\n",
       "      <td>SINGAPORE</td>\n",
       "      <td>1.330005</td>\n",
       "      <td>103.888603</td>\n",
       "      <td>SINGAPORE 409044</td>\n",
       "      <td>SINGAPORE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>745</th>\n",
       "      <td>2017-06-26</td>\n",
       "      <td>SHENG SIONG CIRCUIT ROAD</td>\n",
       "      <td>#01-260,, 18B CIRCUIT RD, SINGAPORE 372018</td>\n",
       "      <td></td>\n",
       "      <td>SINGAPORE</td>\n",
       "      <td>1.329982</td>\n",
       "      <td>103.888608</td>\n",
       "      <td></td>\n",
       "      <td>SINGAPORE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>734</th>\n",
       "      <td>2017-06-24</td>\n",
       "      <td>None</td>\n",
       "      <td>233 PAYA LEBAR RD, SINGAPORE 409044</td>\n",
       "      <td>40904</td>\n",
       "      <td>SINGAPORE</td>\n",
       "      <td>1.329983</td>\n",
       "      <td>103.888558</td>\n",
       "      <td>SINGAPORE 409044</td>\n",
       "      <td>SINGAPORE</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            Date                           name_clean  \\\n",
       "5470  2022-09-16                      HOTEL TARONGETA   \n",
       "5461  2022-09-15                      HOTEL TARONGETA   \n",
       "5453  2022-09-14                        HOTEL GRI-MAR   \n",
       "5443  2022-09-13                        HOTEL GRI-MAR   \n",
       "766   2017-07-02  IBIS STYLES SINGAPORE ON MACPHERSON   \n",
       "759   2017-06-30  IBIS STYLES SINGAPORE ON MACPHERSON   \n",
       "753   2017-06-29                                 None   \n",
       "751   2017-06-27                         LE CRESCENDO   \n",
       "745   2017-06-26             SHENG SIONG CIRCUIT ROAD   \n",
       "734   2017-06-24                                 None   \n",
       "\n",
       "                                          address_clean Zipcode       City  \\\n",
       "5470  PARATGE TARONGETA, S/N, 17488 CADAQUES, GIRONA...     S/N              \n",
       "5461  PARATGE TARONGETA, S/N, 17488 CADAQUES, GIRONA...     S/N              \n",
       "5453     N-260, 165, BAJO, 17490 LA VILA, GIRONA, SPAIN     165              \n",
       "5443     N-260, 165, BAJO, 17490 LA VILA, GIRONA, SPAIN     165              \n",
       "766                 401 MACPHERSON RD, SINGAPORE 368215   36821  SINGAPORE   \n",
       "759                 401 MACPHERSON RD, SINGAPORE 368215   36821  SINGAPORE   \n",
       "753                 233 PAYA LEBAR RD, SINGAPORE 409044   40904  SINGAPORE   \n",
       "751                 233 PAYA LEBAR RD, SINGAPORE 409044   40904  SINGAPORE   \n",
       "745          #01-260,, 18B CIRCUIT RD, SINGAPORE 372018          SINGAPORE   \n",
       "734                 233 PAYA LEBAR RD, SINGAPORE 409044   40904  SINGAPORE   \n",
       "\n",
       "      centerLat   centerLng     Zipcode & town    Country  \n",
       "5470  42.290032    3.273662                S/N      SPAIN  \n",
       "5461  42.290061    3.273581                S/N      SPAIN  \n",
       "5453  42.370081    3.153537                165      SPAIN  \n",
       "5443  42.370103    3.153513                165      SPAIN  \n",
       "766    1.331761  103.879298   SINGAPORE 368215  SINGAPORE  \n",
       "759    1.331767  103.879257   SINGAPORE 368215  SINGAPORE  \n",
       "753    1.329978  103.888539   SINGAPORE 409044  SINGAPORE  \n",
       "751    1.330005  103.888603   SINGAPORE 409044  SINGAPORE  \n",
       "745    1.329982  103.888608                     SINGAPORE  \n",
       "734    1.329983  103.888558   SINGAPORE 409044  SINGAPORE  "
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Objective: I want to identify the place I stayed the most every day thanks to Google Maps data. \n",
    "\n",
    "#Step 1: I exclude work and private home addresses from the data\n",
    "    # This step is not detailed in the public version of the notebook for privacy reasons\n",
    "all_non_personal_data = place_visits_df[(place_visits_df[\"Address_type\"]!=\"WORK\") & (place_visits_df[\"Address_type\"]!=\"HOME\")]\n",
    "\n",
    "#Step 2: I compute for every date the maximum duration\n",
    "#max_day = all_non_personal_data.groupby(\"Date\")[\"duration_minutes\"].max()\n",
    "max_day = all_non_personal_data.groupby(\"Date\")[\"duration_minutes\"].max()\n",
    "max_day.reset_index()\n",
    "max_day = pd.DataFrame(max_day).rename(columns = {\"duration_minutes\" : \"max_dur_day\"})\n",
    "\n",
    "#Step 3: I integrate this maximum duration into the dataframe that contains no private information\n",
    "check1 = max_day.merge(place_visits_df, on = \"Date\")\n",
    "\n",
    "#Step4: I keep only the rows where the duration_minutes is equal to the maximum duration.\n",
    "    # This will give me the name of the places where I spent most of my time during the day (home and work excluded)\n",
    "final_data = check1[[\"Date\", \"name_clean\", \"address_clean\", \n",
    "                    \"Zipcode\", \"City\", \"centerLat\", \"centerLng\",\n",
    "                    \"Zipcode & town\", \"Country\"]][check1[\"duration_minutes\"]== check1[\"max_dur_day\"]]\n",
    "\n",
    "final_data.sort_values([\"Country\", \"Date\"], ascending = False).head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "5f7b7b2f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>counts</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>date</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2016-11-15</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-08-13</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-07-11</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-07-10</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-07-09</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-02-26</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-02-25</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-02-24</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-02-22</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2023-08-06</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1931 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            counts\n",
       "date              \n",
       "2016-11-15       1\n",
       "2021-08-13       1\n",
       "2021-07-11       1\n",
       "2021-07-10       1\n",
       "2021-07-09       1\n",
       "...            ...\n",
       "2019-02-26       1\n",
       "2019-02-25       1\n",
       "2019-02-24       1\n",
       "2019-02-22       1\n",
       "2023-08-06       1\n",
       "\n",
       "[1931 rows x 1 columns]"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Do I have duplicates ? No, it is okay !\n",
    "test = final_data[\"Date\"].value_counts(ascending = False).rename_axis('date').to_frame(\"counts\")\n",
    "test"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "46e439bc",
   "metadata": {},
   "source": [
    "<div style=\"background-color: RGB(0,150,250);\" >\n",
    "<h2 style=\"margin: auto; padding: 20px; color:#fff; \">8. Export final datasets to Excel (csv format)</h2> <a class=\"anchor\" id=\"8\"></a>\n",
    "</div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "b699f1c8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Date</th>\n",
       "      <th>duration_minutes</th>\n",
       "      <th>Sports</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>103</th>\n",
       "      <td>2023-06-01</td>\n",
       "      <td>73.133667</td>\n",
       "      <td>SWIMMING</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>105</th>\n",
       "      <td>2023-06-03</td>\n",
       "      <td>65.649350</td>\n",
       "      <td>SWIMMING</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>110</th>\n",
       "      <td>2023-06-04</td>\n",
       "      <td>84.633233</td>\n",
       "      <td>SWIMMING</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>165</th>\n",
       "      <td>2023-06-18</td>\n",
       "      <td>77.692683</td>\n",
       "      <td>SWIMMING</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>175</th>\n",
       "      <td>2023-06-20</td>\n",
       "      <td>67.239383</td>\n",
       "      <td>SWIMMING</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6212</th>\n",
       "      <td>2022-05-03</td>\n",
       "      <td>86.203783</td>\n",
       "      <td>SWIMMING</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6214</th>\n",
       "      <td>2022-05-04</td>\n",
       "      <td>96.316550</td>\n",
       "      <td>SWIMMING</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6257</th>\n",
       "      <td>2022-05-17</td>\n",
       "      <td>52.993500</td>\n",
       "      <td>SWIMMING</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6262</th>\n",
       "      <td>2022-05-19</td>\n",
       "      <td>98.125817</td>\n",
       "      <td>SWIMMING</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6275</th>\n",
       "      <td>2022-05-24</td>\n",
       "      <td>60.618100</td>\n",
       "      <td>SWIMMING</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>195 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            Date  duration_minutes    Sports\n",
       "103   2023-06-01         73.133667  SWIMMING\n",
       "105   2023-06-03         65.649350  SWIMMING\n",
       "110   2023-06-04         84.633233  SWIMMING\n",
       "165   2023-06-18         77.692683  SWIMMING\n",
       "175   2023-06-20         67.239383  SWIMMING\n",
       "...          ...               ...       ...\n",
       "6212  2022-05-03         86.203783  SWIMMING\n",
       "6214  2022-05-04         96.316550  SWIMMING\n",
       "6257  2022-05-17         52.993500  SWIMMING\n",
       "6262  2022-05-19         98.125817  SWIMMING\n",
       "6275  2022-05-24         60.618100  SWIMMING\n",
       "\n",
       "[195 rows x 3 columns]"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Export the workout data to csv format\n",
    "pd.set_option('display.max_columns', 500)\n",
    "swim_dance_workouts = place_visits_df[~place_visits_df[\"Sports\"].isna()]\n",
    "swim_dance_workouts = swim_dance_workouts[[\"Date\", \"duration_minutes\", \"Sports\"]]\n",
    "swim_dance_workouts.to_csv(\"swim_dance_workouts.csv\", index = False, sep = \";\")\n",
    "swim_dance_workouts"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "9fa68320",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Location day by day\n",
    "final_data = final_data.drop(['name_clean', 'address_clean', 'Zipcode & town'], axis=1)\n",
    "final_data.to_csv(\"final_data.csv\", index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5f010352",
   "metadata": {},
   "source": [
    "<div style=\"background-color: RGB(0,150,250);\" >\n",
    "<h2 style=\"margin: auto; padding: 20px; color:#fff; \">Conclusion and next steps</h2> <a class=\"anchor\" id=\"8\"></a>\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d1eb224c",
   "metadata": {},
   "source": [
    "### Key learnings <a class=\"anchor\" id=\"8.1\"></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4c00f9c6",
   "metadata": {},
   "source": [
    "Thanks to Google Maps History Location data I could recover my swimming and dance workout history data. \n",
    "  \n",
    "The results show that I have not been consistent in taking dance classes because lots of external events made me give up (moving out, transportation system strikes, Covid19). On the contrary, I have pretty stuck to swimming (despite Covid19) and my workouts duration is quite stable over time.\n",
    "\n",
    "From a technical point of view, this project made me work with deep nested JSON files. The learning curve was a bit steep at the beginning but then the data analysis and visualization was easier."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1940df1c",
   "metadata": {},
   "source": [
    "### Next steps <a class=\"anchor\" id=\"8.2\"></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c531cac9",
   "metadata": {},
   "source": [
    "My next steps are the following:\n",
    "- Combine the Samsung Health pedometer data and Google Maps History location data to have the full perspective on my workouts (running, hiking, swimming, dancing)\n",
    "- Combine the Samsung Health pedometer data and Google Maps History location data to provide more context on my pedometer data. The goal is to associate a zipcode / a city to each day I hiked quite a lot.\n",
    "- To further investigate the activitySegments data by focusing on a hike trip I did: where did I go, taking which route, etc?"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8a5f4fe0",
   "metadata": {},
   "source": [
    "Thank you for reading !  \n",
    "\n",
    "<img src=\"https://images.unsplash.com/photo-1461183479101-6c14cd5299c4?ixlib=rb-4.0.3&ixid=MnwxMjA3fDB8MHxwaG90by1wYWdlfHx8fGVufDB8fHx8&auto=format&fit=crop&w=870&q=80\" alt=\"Google Maps image\" width=\"500\" style=\"margin-left:0\"/>  \n",
    "Photo by <a href=\"https://unsplash.com/@stevep4?utm_source=unsplash&utm_medium=referral&utm_content=creditCopyText\">Stephen Monroe</a> on <a href=\"https://unsplash.com/s/photos/route?utm_source=unsplash&utm_medium=referral&utm_content=creditCopyText\">Unsplash</a>\n",
    "  "
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}