09-recsys

1296 erroneous user average 10651 movie average 10533

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Prize: 0.8563; 10% improvement Inherent noise: ???? 2/2/2011 Jure Leskovec, Stanford C246: Mining Massive Datasets accurate 8 [Bellkor Team] Earliest and most popular collaborative filtering method Derive unknown ratings from those of “similar” items (movie-movie variant) A parallel user-user flavor: rely on ratings of like-minded users (not in this talk) 2/2/2011 Jure Leskovec, Stanford C246: Mining Massive Datasets 9 [Bellkor Team] users 1 1 2 1 5 movies 2 4 6 4 2 5 5 4 4 3 6 7 10 11 12 4 4 2 4 2 3 5 3 9 5 4 4 2 3 - unknown rating 2/2/2011 8 5 1 4 1 4 3 2 3 3 Jure Leskovec, Stanford C246: Mining Massive Datasets 1 3 5 2 2 2 3 5 4 - rating between 1 to 5 10 [Bellkor Team] users 1 1 2 1 5 movies 2 4 6 4 2 5 6 5 4 4 3 5 ? 1 4 1 4 3 2 3 3 7 9 10 11 12 5 4 4 2 4 2 3 5 3 8 4 4 2 3 1 3 5 2 2 2 3 5 4 - estimate rating of movie 1 by user 5 2/2/2011 Jure Leskovec, Stanford C246: Mining Massive Datasets 11 [Bellkor Team] users 1 1 2 1 5 movies 2 4 6 4 2 5 6 5 4 4 3 5 ? 1 4 1 4 3 2 3 3 7 9 10 11 12 5 4 4 2 4 2 3 5 3 8 4 4 2 3 1 3 5 2 2 2 3 5 4 Neighbor selection: Identify movies similar to 1, rated by user 5 2/2/2011 Jure Leskovec, Stanford C246: Mining Massive Datasets 12 [Bellkor Team] users 1 1 2 1 5 movies 2 4 6 4 2 5 5 6 ? 5 4 1 4 4 1 4 3 2 3 3 7 3 9 10 11 12 5 4 4 2 4 2 3 5 3 8 4 4 2 3 1 3 5 2 2 2 3 5 4 Compute similarity weights: s13=0.2, s16=0.3 2/2/2011 Jure Leskovec, Stanford C246: Mining Massive Datasets 13 [Bellkor Team] users 1 1 2 1 5 movies 2 4 6 4 2 5 6 4 4 3 5 7 8 10 11 12 4 4 2 4 2 3 5 3 9 5 2.6 5 1 4 1 4 3 2 3 3 4 4 2 3 1 3 5 2 2 2 3 5 4 Predict by taking weighted average: (0.2*2+0.3*3)/(0.2+0.3)=2.6 2/2/2011 Jure Leskovec, Stanford C246: Mining Massive Datasets 14 Intuitive No substantial preprocessing is required Easy to explain reasoning behind a recommendation Accurate? 2/2/2011 Jure Leskovec, Stanford C246: Mining Massive Datasets 15 [Bellkor Team] Global average: 1.1296 erroneous User average: 1.0651 Movie average: 1.0533 0.96 Cinematch: 0.9514 0.9...
View Full Document

This document was uploaded on 02/26/2014 for the course CS 246 at Stanford.

Ask a homework question - tutors are online