Novel Boosting Frameworks to Improve the Performance of Collaborative Filtering

Xiaotian Jiang, Zhendong Niu, Jiamin Guo, Ghulam Mustafa, Zihan Lin, Baomi Chen, Qian Zhou
Proceedings of the 5th Asian Conference on Machine Learning, PMLR 29:87-99, 2013.

Abstract

Recommender systems are often based on collaborative filtering. Previous researches on collaborative filtering mainly focus on one single recommender or formulating hybrid with different approaches. In consideration of the problems of sparsity, recommender error rate, sample weight update, and potential, we adapt AdaBoost and propose two novel boosting frameworks for collaborative filtering. Each of the frameworks combines multiple homogeneous recommenders, which are based on the same collaborative filtering algorithm with different sample weights. We use seven popular collaborative filtering algorithms to evaluate the two frameworks with two MovieLens datasets of different scale. Experimental result shows the proposed frameworks improve the performance of collaborative filtering.

Cite this Paper


BibTeX
@InProceedings{pmlr-v29-Jiang13, title = {Novel Boosting Frameworks to Improve the Performance of Collaborative Filtering}, author = {Jiang, Xiaotian and Niu, Zhendong and Guo, Jiamin and Mustafa, Ghulam and Lin, Zihan and Chen, Baomi and Zhou, Qian}, booktitle = {Proceedings of the 5th Asian Conference on Machine Learning}, pages = {87--99}, year = {2013}, editor = {Ong, Cheng Soon and Ho, Tu Bao}, volume = {29}, series = {Proceedings of Machine Learning Research}, address = {Australian National University, Canberra, Australia}, month = {13--15 Nov}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v29/Jiang13.pdf}, url = {https://proceedings.mlr.press/v29/Jiang13.html}, abstract = {Recommender systems are often based on collaborative filtering. Previous researches on collaborative filtering mainly focus on one single recommender or formulating hybrid with different approaches. In consideration of the problems of sparsity, recommender error rate, sample weight update, and potential, we adapt AdaBoost and propose two novel boosting frameworks for collaborative filtering. Each of the frameworks combines multiple homogeneous recommenders, which are based on the same collaborative filtering algorithm with different sample weights. We use seven popular collaborative filtering algorithms to evaluate the two frameworks with two MovieLens datasets of different scale. Experimental result shows the proposed frameworks improve the performance of collaborative filtering.} }
Endnote
%0 Conference Paper %T Novel Boosting Frameworks to Improve the Performance of Collaborative Filtering %A Xiaotian Jiang %A Zhendong Niu %A Jiamin Guo %A Ghulam Mustafa %A Zihan Lin %A Baomi Chen %A Qian Zhou %B Proceedings of the 5th Asian Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2013 %E Cheng Soon Ong %E Tu Bao Ho %F pmlr-v29-Jiang13 %I PMLR %P 87--99 %U https://proceedings.mlr.press/v29/Jiang13.html %V 29 %X Recommender systems are often based on collaborative filtering. Previous researches on collaborative filtering mainly focus on one single recommender or formulating hybrid with different approaches. In consideration of the problems of sparsity, recommender error rate, sample weight update, and potential, we adapt AdaBoost and propose two novel boosting frameworks for collaborative filtering. Each of the frameworks combines multiple homogeneous recommenders, which are based on the same collaborative filtering algorithm with different sample weights. We use seven popular collaborative filtering algorithms to evaluate the two frameworks with two MovieLens datasets of different scale. Experimental result shows the proposed frameworks improve the performance of collaborative filtering.
RIS
TY - CPAPER TI - Novel Boosting Frameworks to Improve the Performance of Collaborative Filtering AU - Xiaotian Jiang AU - Zhendong Niu AU - Jiamin Guo AU - Ghulam Mustafa AU - Zihan Lin AU - Baomi Chen AU - Qian Zhou BT - Proceedings of the 5th Asian Conference on Machine Learning DA - 2013/10/21 ED - Cheng Soon Ong ED - Tu Bao Ho ID - pmlr-v29-Jiang13 PB - PMLR DP - Proceedings of Machine Learning Research VL - 29 SP - 87 EP - 99 L1 - http://proceedings.mlr.press/v29/Jiang13.pdf UR - https://proceedings.mlr.press/v29/Jiang13.html AB - Recommender systems are often based on collaborative filtering. Previous researches on collaborative filtering mainly focus on one single recommender or formulating hybrid with different approaches. In consideration of the problems of sparsity, recommender error rate, sample weight update, and potential, we adapt AdaBoost and propose two novel boosting frameworks for collaborative filtering. Each of the frameworks combines multiple homogeneous recommenders, which are based on the same collaborative filtering algorithm with different sample weights. We use seven popular collaborative filtering algorithms to evaluate the two frameworks with two MovieLens datasets of different scale. Experimental result shows the proposed frameworks improve the performance of collaborative filtering. ER -
APA
Jiang, X., Niu, Z., Guo, J., Mustafa, G., Lin, Z., Chen, B. & Zhou, Q.. (2013). Novel Boosting Frameworks to Improve the Performance of Collaborative Filtering. Proceedings of the 5th Asian Conference on Machine Learning, in Proceedings of Machine Learning Research 29:87-99 Available from https://proceedings.mlr.press/v29/Jiang13.html.

Related Material