Altair RISE

A program to recognize and reward our most engaged community members

Nominate Yourself Now!

Collaborative filtering: multiple items are rated multiple times by each user

Would it be possible to model as a recommendation problem the selection of an item that can be selected (and rated) by the same user many times? I can't find references of previous work.

For example User1 can select item A (and rate it) then item A again (and give it a different rate) and then again item A can be selected n times.. then the same for item B, but item A could be again selected at any time point, and that could happen for any of the A, B, C... Z items. Could this be modelled somehow as a recommender system?

Find more posts tagged with

AI Studio

Filtering

Accepted answers

All comments

juanm_encinas

I'd say that the main point here is to decide what to do with multiple ratings of a same item by one same user. Do you pick the last rating? The average of all ratings? The maximum rating? In the last two cases I'd aggregate ratings by user first (either using average or maximum aggregation functions) to build my recommendation table. And then proceed with the recommender system.

JEdward

I'd be tempted to experiment with weighting the average values based on most recent review (to stop people trying to rate several times to game the system, but also reflecting that people can change their minds).

For example give a worth of 1 to the most recent review, then 0.9,0.8,0.7 etc for earlier ones.

So, for example if someone rates a Product A 5 times:
Rating
5* most recent
3
2
2
1
4* least recent

Taking an average of the reviews would weight it at 2.6 which doesn't really reflect the high recent review, but if you weight them as above then the average becomes 3.8 much closer to the most recent review.