9,000 rows · 7 columns
Movie ratings
Nine thousand user ratings over 260 films, with genres, release years, and the long-tail popularity distribution that makes naive recommenders look better than they are.
What is in it, and what is wrong with it
Ratings are missing-not-at-random: people mostly rate films they chose to watch, which is why an average rating is a popularity measure as much as a quality measure. Twelve films have fewer than five ratings — those are where a "top rated" leaderboard goes wrong without a shrinkage prior.
Schema
| Column | Type | Description |
|---|---|---|
| user_id | string | Rater identifier. |
| movie_id | string | Film identifier. |
| title | string | Film title. |
| genre | category | Primary genre. |
| release_year | integer | Year of release. |
| rating | float | Star rating, 0.5–5.0 in half steps. |
| rated_at | timestamp | When the rating was submitted. |
First twelve rows
| user_id | movie_id | title | genre | release_year | rating | rated_at |
|---|---|---|---|---|---|---|
| U9826 | M1244 | Last Harbour | comedy | 2022 | 3 | 2023-01-01 00:22:37 |
| U10084 | M1100 | Electric Harbour II | comedy | 1972 | 4 | 2023-01-01 12:37:28 |
| U9848 | M1027 | Electric Harbour | comedy | 2006 | 4.5 | 2023-01-01 16:44:49 |
| U9217 | M1013 | Silent Signal | drama | 2019 | 3 | 2023-01-01 17:28:05 |
| U9491 | M1019 | Second Lantern | comedy | 2017 | 4 | 2023-01-01 18:03:19 |
| U9962 | M1000 | Iron Country | romance | 2012 | 5 | 2023-01-01 18:26:21 |
| U9947 | M1000 | Iron Country | romance | 2012 | 3 | 2023-01-01 22:46:52 |
| U9422 | M1002 | Pale Meridian | thriller | 2008 | 3 | 2023-01-01 22:47:15 |
| U9825 | M1243 | Crimson Archive | romance | 1975 | 4 | 2023-01-02 00:47:38 |
| U9378 | M1004 | Crimson Harbour | comedy | 2000 | 4 | 2023-01-02 01:08:31 |
| U9085 | M1097 | Broken Machine | romance | 2017 | 2 | 2023-01-02 02:27:11 |
| U9257 | M1087 | Second Orchard | drama | 1978 | 3.5 | 2023-01-02 04:27:17 |