This is a full fledged whatsapp analyser intended to work on any conversation. Apart from what dsquareIndia has done, I have added a word-cloud generator twoards the end of the notebook. Here I have used a Whatsapp chat of mine with my friend to generate the wordCloud that we see at the end of the file.
From dsquareIndia's readme: My aim is to build an app which can be used by anyone in any country typing in any language. This is just the initial release wherein I have built the following features:
- Most used words in a conversation
- Peak detection on the word-frequency graph (just for the kicks. Doesn't accomplish anything more than the above feature.)
- The most and least talkative people in a group. Create a pie chart of individual % contribution to messages.
- An activity map to analyse during what time of the day is the group most active.
- Most frequently used words by individual participants.
- Conversation trends over full timeline.
- (Typing style similarity. Added as a PR. Still have to improve the way I show it.)
Note that this project is far from complete. I would love to receive new ideas and improvements or bug fixes/reports. Please report any problem you encounter while using this script.
======================
- Email the conversation to yourself (options->more->email chat or simply long press the desired group and then email chat) without media.
- Save this textfile in the same folder as this scipt (or enter full file path in the next step.)
- You will be prompted to enter the name of the conversation with extension. Enter the name and see the results.
- The results are best viewed in an ipython notebook. I used Jupyter IPython Notebook.
======================