Social Media Mining on Twitter Data
=> Requirements
- Anaconda Python Distribution - Version 3.6 https://www.anaconda.com/download/
- Below are the Python libraries needed to be installed. Please use command pip install "library name" using Anaconda Command Prompt
- bokeh
- nltk
- numpy
- sys
- operator
- collections
- copy
- re
- PIL
- wordcloud
- tweepy
- pandas
- geopy
- json
- os
- matplotlib
- Please keep the "SMMProcessModule.ipynb", "SMM - Term Project - Main.ipynb", Catogeries.csv, twittermask.jpg in the same folder. Make sure to have other files too if running the application for the existing data set
=> How to
-
Unzip the folder in the location accessible by Jupyter Notebook
-
Open the "SMM - Term Project - Main.ipynb" and run all the cell from top to bottom
-
After running the last cell, there will be 2 tabs - one shows list of already collected data for which one can see the visualizations and the second tab takes new input for which the process can be run
-
As this will the first time the application will be running, there wont be any data for first tab. So please use the second tab to generate the results. (It may take upto 30 mins as the data collection is involved for each input entered"
-
Once the data is generated and results produced, it will be available for the first tab in dropdownlist for later use
-
We are already providing some data, to have the dropdown menu pre-populated