KDnuggets Home » News » 2012 » Dec » Publications » ebook: Predicting movie ratings and recommender systems

ebook: Predicting movie ratings and recommender systems


 
  
< Previous post Next post >
Arek Paterek was among top competitors in Netflix Prize. This 195-page monograph includes a comprehensive summary of the Netflix Prize, detailed description of his top-50 Netflix Prize solution predicting movie ratings, and much more.


ebook: Predicting movie ratings and recommender systemsarek-paterek.com/book/ is a new 195-page monograph by A. Paterek.

It extends the earlier 2007 four-page publication (in KDD Cup 2007 workshop) "Improving singular value decomposition for collaborative filtering", now cited over 200 times.

The author placed 43rd in the Netflix Prize, earlier reaching 3-5th place in the first year of the competition, and for a brief moment reaching 2nd place in a two-person team, one day before the first Progress Prize in 2007, hopefully scaring a bit the top teams.

He developed the recommender systems in Movie Galaxy (arek-paterek.com/movie-galaxy/) and in 5000 Best Movies and TV Series (5000best.com/movies/).

Fig. 30 Movie Features

The e-book is an attempt to extensively summarize and comment the Netflix Prize solutions, and discusses how to adapt the resulting prediction algorithms to produce good quality personalized recommendations.

The e-book is available at arek-paterek.com/book/



< Previous post Next post >

KDnuggets Home » News » 2012 » Dec » Publications » ebook: Predicting movie ratings and recommender systems