A LITTLE TASTE. A BETTER WATCHLIST.
Your next
series.
Start with shows you know.
Find something you’ll love.
EARLY EXPERIMENT - Amazon reviews grouped into TV series. Coverage and title matching are imperfect.
What have you seen?
Add a few favorites and a few misses. Three or more is a great start.
Type a title, then select a show from the dropdown.
Like 0
The ones you’d watch again.
Dislike 0
The ones that weren’t for you.
Your picks stay in this browser. No account needed.
Made for your taste.
Your recommendations will appear here.
Good shows start with you.
Add a show to either list and we’ll find your next watch. Keep adding picks to fine-tune the results.
About this first TV experiment
This model learns from the Amazon Reviews 2023 Movies & TV dataset. We identify series using TV categories and season labels, merge matching seasons and editions, and average each person's ratings into one rating per series. The catalog includes only series with at least 50 reviewers after filtering for people who reviewed at least three series.
The five closest liked shows contribute similarity weights to each match score; the five closest disliked shows subtract theirs. We sum these weights, without averaging or sorting by the number of contributors. Expand any recommendation to inspect the contributions. Recommendations exclude every show you entered. You can browse up to 500 results.
Experimental coverage: Amazon reviews can describe disc quality or delivery as well as the show itself. Automatic title matching can miss shows or confuse remakes. Recent streaming-only series may be absent; the source data ends in September 2023. This is an initial research prototype, not a complete TV catalog. TV and movie preference lists are saved separately in your browser.
Source: Hou et al. (2024), Bridging Language and Items for Retrieval and Recommendation, McAuley Lab, UC San Diego. No affiliation with or endorsement by Amazon or the dataset authors. Training and validation results · Series/product mapping · Attribution · Source code