This notebook was built to analyze Whatsapp conversations using the steps below:
- Step 1: Detecting {Date} and {Time} tokens
- Step 2: Detecting the {Author} token
- Step 3: Extracting and Combining tokens
- Step 4: Parsing the entire file and handling Multi-Line Messages
For further steps, we need to perform Exploratory data analysis (EDA)
- Step 5: Performing EDA for analyzing chat data
- Step 6: Overall statistics of WhatsApp chat including Total number of messages, media messages(Omitted) & Total number of URLs
- Step 7: Extracting basic statistics for each Author (user)
- Step 8: Word cloud of most used words in chat
- Step 9: Total number of messages sent by each user
- Step 10: Total messages sent on each day of the week
- Step 11: Most active author of the chat
- Step 12: Most active day in a week
In next steps, Time series analysis will be performed on chat data
- Step 13: Time whenever the chat was highly active
- Step 14: Date on which the chat was highly active
- Step 15: Converting 12-hour formate to 24 hours will help us for better analysis
- Step 16: Most suitable hour of the day whenever there will be more chances of getting a response from user
Project visulizations https://github.com/KashmalaJamshaid/NLP-implementation-on-whastapp-chats-using-python/commit/cdb2f4faf7e8891f00a1ffa9cb46497ac0202bd2#commitcomment-53149178