Technology

Netflix Prize: Best Rated and Most Rated Movies

Posted on by rkischuk in Technology

I’m currenty intrigured by the Netflix Prize – if anyone can improve their score prediction engine by 10%, they win $1 million. While we sliced a lot of data for predictive modeling at Proficient, I don’t expect to have the expertise to solve it, but I love messing with data, so in the mean time, I’m playing with their files of 100 million ratings – 2 GB uncompressed, that they provide for contest participants to train and test their algorithms against.

The first interesting data I’ve derived from this dataset is the highest rated and most frequently rated movies.

The best rated – (score), Dataset average 3.6043

  1. Lord of the Rings: The Return of the King: Extended Edition – 4.72327
  2. The Lord of the Rings: The Fellowship of the Ring: Extended Edition – 4.71661
  3. Lord of the Rings: The Two Towers: Extended Edition – 4.70261
  4. Lost: Season 1 – 4.67099
  5. Battlestar Galactica: Season 1 – 4.63881
  6. Fullmetal Alchemist – 4.60502
  7. Trailer Park Boys: Season 4 – 4.6 (
  8. Trailer Park Boys: Season 3 – 4.6 (
  9. Tenchi Muyo! Ryo Ohki – 4.59551 (
  10. The Shawshank Redemption: Special Edition – 4.59338
  11. Veronica Mars: Season 1 – 4.59208
  12. Ghost in the Shell: Stand Alone Complex: 2nd Gig – 4.58636
  13. Arrested Development: Season 2 – 4.58239
  14. The Simpsons: Season 6 – 4.5813
  15. Inu-Yasha – 4.55443
  16. Lord of the Rings: The Return of the King: Extended Edition: Bonus Material – 4.552
  17. Lord of the Rings: The Return of the King – 4.54512
  18. Star Wars: Episode V: The Empire Strikes Back – 4.5437
  19. The Simpsons: Season 5 – 4.54256
  20. Fruits Basket – 4.53891

The Most Rated – (number of ratings), 100,480,507 total ratings

  1. Miss Congeniality – 232944
  2. Independence Day – 216596
  3. The Patriot – 200832
  4. The Day After Tomorrow – 196397
  5. Pirates of the Caribbean: The Curse of the Black Pearl – 193941
  6. Pretty Woman – 193295
  7. Forrest Gump – 181508
  8. The Green Mile – 181426
  9. Con Air – 178068
  10. Twister – 177556
  11. Sweet Home Alabama – 176539
  12. Pearl Harbor – 173596
  13. Armageddon – 171991
  14. The Rock – 164792
  15. What Women Want – 162597
  16. Bruce Almighty – 160454
  17. Ocean’s Eleven – 160326
  18. The Bourne Identity – 158601
  19. The Italian Job – 156183
  20. I, Robot – 155714

If there are other interesting stats you’d like derived, or you have fancy ideas on how to best predict a user’s rating, drop me an email or comment.

Advertisement