TY - GEN

T1 - The sample complexity of online one-class collaborative filtering

AU - Hecke, Reinhard

AU - Ramchandran, Kannan

N1 - Publisher Copyright:
Copyright 2017 t by the author(s).

PY - 2017

Y1 - 2017

N2 - We consider the online one-class collaborative filtering (CF) problem that consists of recommending items to users over time in an online fashion based on positive ratings only. This problem arises when users respond only occasionally to a recommendation with a positive rating, and never with a negative one. We study the impact of the probability of a user responding to a recommendation, pi, on the sample complexity, i.e., the number of ratings required to make 'good' recommendations, and ask whether receiving positive and negative ratings, instead of positive ratings only, improves the sample complexity. Both questions arise in the design of recommender systems. We introduce a simple probabilistic user model, and analyze the performance of an online user-based CF algorithm. We prove that after an initial cold start phase, where recommendations are invested in exploring the user's preferences, this algorithm makes-up to a fraction of the recommendations required for updating the user's preferences-perfect recommendations. The number of ratings required for the cold start phase is nearly proportional to 1/pf, and that for updating the user's preferences is essentially independent of pf. As a consequence we find that, receiving positive and negative ratings instead of only positive ones improves the number of ratings required for initial exploration by a factor of 1/pf, which can be significant.

AB - We consider the online one-class collaborative filtering (CF) problem that consists of recommending items to users over time in an online fashion based on positive ratings only. This problem arises when users respond only occasionally to a recommendation with a positive rating, and never with a negative one. We study the impact of the probability of a user responding to a recommendation, pi, on the sample complexity, i.e., the number of ratings required to make 'good' recommendations, and ask whether receiving positive and negative ratings, instead of positive ratings only, improves the sample complexity. Both questions arise in the design of recommender systems. We introduce a simple probabilistic user model, and analyze the performance of an online user-based CF algorithm. We prove that after an initial cold start phase, where recommendations are invested in exploring the user's preferences, this algorithm makes-up to a fraction of the recommendations required for updating the user's preferences-perfect recommendations. The number of ratings required for the cold start phase is nearly proportional to 1/pf, and that for updating the user's preferences is essentially independent of pf. As a consequence we find that, receiving positive and negative ratings instead of only positive ones improves the number of ratings required for initial exploration by a factor of 1/pf, which can be significant.

UR - http://www.scopus.com/inward/record.url?scp=85048446911&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:85048446911

T3 - 34th International Conference on Machine Learning, ICML 2017

SP - 2299

EP - 2315

BT - 34th International Conference on Machine Learning, ICML 2017

PB - International Machine Learning Society (IMLS)

T2 - 34th International Conference on Machine Learning, ICML 2017

Y2 - 6 August 2017 through 11 August 2017

ER -