Assume I have a large data of ecommerce website sessions with user id key (a user can have multiple sessions with random time between them). The data is on S3 in json gzipped format. On some sessions there are purchases.
I want to predict the chance that a user will purchase in the next session for a given session data.
To train it I need to add for each session a data point about the chance that there will be purchase in the next session.
What is the best tool/way to prepare such training data set? I believe I needs to scan the sessions from end to start and remember when was the last purchase time for each user id to update the earlier records of each user. How can this be accomplished Using AWS tools (Glue, Batch etc..)