This is a super exciting data set and I hope you can get more out of it: is it actually worth it to prioritize the features people are asking for? how long does it take to recover from a drop, can this be modeled somehow? how come some apps receive inconsistent emotion? etc
I would be very happy to receive more replies like yours to come up with new ideas and different perspectives on how to look at the data, and to prioritize which questions to answer.
- Is there any seasonality to review volume and/or sentiment (day of week, time of month/year/etc) that holds across many apps?
- Are there interesting differences in sentiment mean+variance by app category? For example, are utility app reviews more polarized than games?
> - Are there interesting differences in sentiment mean+variance by app category? For example, are utility app reviews more polarized than games?
These are answered in the white paper: https://arxiv.org/pdf/1703.02256.pdf