[General Statistics] Analysis daily behaviour

kiTcuT-Nussi commented 3 years ago

Try to meassure and analyze daily behaviour. Possible idea: Heatmap representing the "usual" daily activity of the senders for each day and every hour

get top words of sections with high chat activity
get top words of mondays
get top words of tuesdays
...
get most words in the moring at 09, at 10, at 11....

Map of all days as bubbles and most used words / topics inside bubble and distance between words maps to their co-occurancy their "purity" in terms of occurence on a single day, e.g. "beer" only appears on saturdays. "Grocery shopping" only is used on tuesdays, and so on...

kiTcuT-Nussi commented 3 years ago

generate heatmap over daily activities with plot_weekday_activity(chat_df)

kiTcuT-Nussi commented 3 years ago

Word frequencys and appearaces for weekdays in general is not really interesting, because these words are kind of meaningless and generic, even if filtered for nouns only.

kiTcuT-Nussi / whatsapp_chat_analyzer

[General Statistics] Analysis daily behaviour #9