View
770
Download
2
Category
Preview:
DESCRIPTION
Citation preview
2. Outline
3.
4. Tutorial: Neighbor Method
Related Rocky Balboa Hasnt rated it guess 5 stars? Rocky Rates it 5 Stars Rocky IV Rates it 5 stars Suppose the attributes aremovie titles and a users ratings of those movies.The task is to predict what that user will rate a new movie. Pretty Woman Rates it 2 Stars 5. Relatedness
The catch is to define related - Sarwar, et al. Item-based collaborativefiltering recommendation algorithms.2001. How to begin to understand a relatedness measure? 1. Off the shelf measures Pearson Correlation 2. Tailor measure to dataset 6. Visualization of Relatedness Measure
Fruchterman & Reingold.Graph drawing by Force Directed Placement.1991. 0.8 0.5 0.6 Proximity is interpreted as relatedness
7. Visualization of Relatedness Measure Whats the big cluster in the center? 8. Assembling the Model
This is similar to the approaches reported by Amazon in 2003, and Tivo in 2004. - Sarwar, et al. Item-based collaborativefiltering recommendation algorithms.2001. Training Examples Relatedness / Similarity 9.
Ensemble Learning in Practice:A Look at the Netflix Prize October 2006-present 10.
From the Internet Archive. 11. However, improvement slowed and techniques became more sophisticated Bennett and Lanning. KDCup 2007. 12. Bennett and Lanning. KDCup 2007. Techniques used 13. Thanks to Paul Harrison's collaboration, a simple mix of our solutions improved our result from 6.31 to 6.75 Rookies (35) 14. My approach is tocombine the results of many methods(also two-way interactions between them) using linear regression on the test set. The best method in my ensemble is regularized SVD with biases, post processed with kernel ridge regression Arek Paterek (15) http://rainbow.mimuw.edu.pl/~ap/ap_kdd.pdf 15. When the predictions ofmultipleRBM models andmultipleSVD models are linearly combined, we achieve an error rate that is well over 6% better than the score of Netflixs own system. U of Toronto (13) http://www.cs.toronto.edu/~rsalakhu/papers/rbmcf.pdf 16. Gravity (3) home.mit.bme.hu/~gtakacs/download/gravity.pdf 17. Predictive accuracy is substantially improved when blending multiple predictors. Our experience is that most efforts should be concentrated in deriving substantially different approaches, rather than refining a single technique. Consequently, oursolution is an ensemble of many methods. Our final solution (RMSE=0.8712) consists of blending 107 individual results. BellKor (2) http://www.research.att.com/~volinsky/netflix/ProgressPrize2007BellKorSolution.pdf 18. Our common team blends the result of team Gravity and team Dinosaur Planet. Might have guessed from the name When Gravity andDinosaurs Unite (1) 19. Why combine models?
20. A Reflection
SeeDomingos, P. Occams two razors: the sharp and the blunt.KDD.1998. 21. Achieving Diversity
22. Achieving Diversity
Ratings Actors Genres Classifier A Classifier B Classifier C + Predictions + Classifier A Classifier B Classifier C + Predictions + Training Examples Training Examples 23. Two Particular Strategies
24. Bagging Diversity
25. Bagging Algorithm
26. Boosting Incrementally create models using selectively using training examples based on some distribution. 27. AdaBoost (Adaptive Boosting) Algorithm
28. AdaBoost Cont.
29. Recap
30. Further Information
Posted on www.ABeautifulWWW.com
Recommended