180 likes | 202 Views
Methods and Metrics for Cold-Start Recommendations. Andrew I. Schein, Alexandrin Popescul, Lyle H. Ungar and David M. Pennock. Introduction. Recommender System suggest items of interest based on previous explicit/implicit ratings Ratings explicit rating by user feedback implicit rating
E N D
Methods and Metrics for Cold-Start Recommendations Andrew I. Schein, Alexandrin Popescul, Lyle H. Ungar and David M. Pennock
Introduction • Recommender System • suggest items of interest based on previous explicit/implicit ratings • Ratings • explicit rating • by user feedback • implicit rating • by user action (purchase, refer, click links ... ) BI lab.
Collaborative Filtering • by recommendations on community preferences • Content-based Filtering • matching user (id, query, demographic info.) with items • Cold-Start problem • when no one rated a specific item yet. BI lab.
In this paper, • evaluation of 2 algo. in cold-start setup • propose CROC curve metric BI lab.
Background and Related Work • early recommenders • pure collaborative • similarity-weighted average (memory-based algo.) • hard-clustering users • hard-clustering users and items • soft-clustering users and items • singular value decomposition • inferring item-item similarities • probabilistic modeling • list-ranking BI lab.
Recently, • hybrid recommender systems • combining collaborative and content information • Evaluations • MAE, ROC, ranked list metrics, variants of precision/recall statistics BI lab.
3. The Two-Way Aspect Model • designed for contingency table smoothing BI lab.
3.1 Pure Collaborative Filtering Model • Aspect model • encodes a probability distribution over each person/movie pair. • have a hidden or latent cause z that motivates person p to watch movie m. • m is assumed independent of p given knowledge of z. • P(p,m) : smoothed estimates of the probability distribution of the contingency table BI lab.
the parameters are calculated using EM algorithm • # of latent vars : using performance training data • learning by maximizing P(p,m) • prediction by • Folding In • in person/actor model, we must create a new movie object out of the set of actors that appear in that movie BI lab.
4. Naive Bayes Recommender • bag-of-words naive Bayes classifier applied to person/actor data • pure content-based method • Rating prediction : BI lab.
5.Testing Methodology • MovieLens data set • Each person rates at least twenty movies • rating : 1-5 • actor & director info. from www.imdb.com • test set of 19,192 out of 312,133 • testing modes • implicit rating prediction • rating prediction • rating imputation BI lab.
6. Evaluation Metrics • Receiver Operator Characteristic (ROC) curve [Herlocker] • showing hit/miss rates for different classification thresholds • sensitivity (hit rate) • percentage of all positive values found above some threshold • 1-specificity (miss rate) • the fraction of all negative values found above some threshold BI lab.
Global ROC (GROC) curve • when allowed to recommend more to some users than others • 1. Order pred(pi, mj) by magnitude. • 2. Pick n, hit/miss rate. • Customer ROC (CROC) curve • when constrained to recommend the same number to each user. BI lab.
7. Results BI lab.