A UCB-Like Strategy of Collaborative Filtering

Atsuyoshi Nakamura
Proceedings of the Sixth Asian Conference on Machine Learning, PMLR 39:315-329, 2015.

Abstract

We consider a direct mail problem in which a system repeats the following process every day during some period: select a set of user-item pairs (u,i), send a recommendation mail of item i to user u for each selected pair (u,i), and receive a response from each user. We assume that each response can be obtained before the next process and through the response, the system can know the user’s evaluation of the recommended item directly or indirectly. Each pair (u,i) can be selected at most once during the period. If the total number of selections is very small compared to the number of entries in the whole user-item matrix, what selection strategy should be used to maximize the total sum of users’ evaluations during the period? We consider a UCB-like strategy for this problem, and show two methods using the strategy. The effectiveness of our methods are demonstrated by experiments using synthetic and real datasets.

Cite this Paper


BibTeX
@InProceedings{pmlr-v39-nakamura14, title = {A {UCB}-Like Strategy of Collaborative Filtering}, author = {Nakamura, Atsuyoshi}, booktitle = {Proceedings of the Sixth Asian Conference on Machine Learning}, pages = {315--329}, year = {2015}, editor = {Phung, Dinh and Li, Hang}, volume = {39}, series = {Proceedings of Machine Learning Research}, address = {Nha Trang City, Vietnam}, month = {26--28 Nov}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v39/nakamura14.pdf}, url = {https://proceedings.mlr.press/v39/nakamura14.html}, abstract = {We consider a direct mail problem in which a system repeats the following process every day during some period: select a set of user-item pairs (u,i), send a recommendation mail of item i to user u for each selected pair (u,i), and receive a response from each user. We assume that each response can be obtained before the next process and through the response, the system can know the user’s evaluation of the recommended item directly or indirectly. Each pair (u,i) can be selected at most once during the period. If the total number of selections is very small compared to the number of entries in the whole user-item matrix, what selection strategy should be used to maximize the total sum of users’ evaluations during the period? We consider a UCB-like strategy for this problem, and show two methods using the strategy. The effectiveness of our methods are demonstrated by experiments using synthetic and real datasets.} }
Endnote
%0 Conference Paper %T A UCB-Like Strategy of Collaborative Filtering %A Atsuyoshi Nakamura %B Proceedings of the Sixth Asian Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2015 %E Dinh Phung %E Hang Li %F pmlr-v39-nakamura14 %I PMLR %P 315--329 %U https://proceedings.mlr.press/v39/nakamura14.html %V 39 %X We consider a direct mail problem in which a system repeats the following process every day during some period: select a set of user-item pairs (u,i), send a recommendation mail of item i to user u for each selected pair (u,i), and receive a response from each user. We assume that each response can be obtained before the next process and through the response, the system can know the user’s evaluation of the recommended item directly or indirectly. Each pair (u,i) can be selected at most once during the period. If the total number of selections is very small compared to the number of entries in the whole user-item matrix, what selection strategy should be used to maximize the total sum of users’ evaluations during the period? We consider a UCB-like strategy for this problem, and show two methods using the strategy. The effectiveness of our methods are demonstrated by experiments using synthetic and real datasets.
RIS
TY - CPAPER TI - A UCB-Like Strategy of Collaborative Filtering AU - Atsuyoshi Nakamura BT - Proceedings of the Sixth Asian Conference on Machine Learning DA - 2015/02/16 ED - Dinh Phung ED - Hang Li ID - pmlr-v39-nakamura14 PB - PMLR DP - Proceedings of Machine Learning Research VL - 39 SP - 315 EP - 329 L1 - http://proceedings.mlr.press/v39/nakamura14.pdf UR - https://proceedings.mlr.press/v39/nakamura14.html AB - We consider a direct mail problem in which a system repeats the following process every day during some period: select a set of user-item pairs (u,i), send a recommendation mail of item i to user u for each selected pair (u,i), and receive a response from each user. We assume that each response can be obtained before the next process and through the response, the system can know the user’s evaluation of the recommended item directly or indirectly. Each pair (u,i) can be selected at most once during the period. If the total number of selections is very small compared to the number of entries in the whole user-item matrix, what selection strategy should be used to maximize the total sum of users’ evaluations during the period? We consider a UCB-like strategy for this problem, and show two methods using the strategy. The effectiveness of our methods are demonstrated by experiments using synthetic and real datasets. ER -
APA
Nakamura, A.. (2015). A UCB-Like Strategy of Collaborative Filtering. Proceedings of the Sixth Asian Conference on Machine Learning, in Proceedings of Machine Learning Research 39:315-329 Available from https://proceedings.mlr.press/v39/nakamura14.html.

Related Material