TY - JOUR
T1 - Extended Latent Class Models for Collaborative Recommendation
AU - Cheung, Kwok Wai
AU - Tsui, Kwok Ching
AU - Liu, Jiming
N1 - Funding Information:
Manuscript received February 27, 2001; revised September 27, 2002 and July 10, 2003. This work was jointly supported by Hong Kong Baptist University via Faculty Research under Grant FRG/99-00/II-36P and by RGC Grant HKBU/2090/01E.
PY - 2004/1
Y1 - 2004/1
N2 - With the advent of the World Wide Web, providing just-in-time personalized product recommendations to customers now becomes possible. Collaborative recommender systems utilize correlation between customer preference ratings to identify "like-minded" customers and predict their product preference. One factor determining the success of the recommender systems is the prediction accuracy, which in many cases is limited by lacking adequate ratings (the sparsity problem). Recently, the use of latent class model (LCM) has been proposed to alleviate this problem. In this paper, we first study how the LCM can be extended to handle customers and products outside the training set. In addition, we propose the use of a pair of LCMs (called dual latent class model-DLCM), instead of a single LCM, to model customers' likes and dislikes separately for enhancing the prediction accuracy. Experimental results based on the EachMovie dataset show that DLCM outperforms both LCM and the conventional correlation-based method when the available ratings are sparse.
AB - With the advent of the World Wide Web, providing just-in-time personalized product recommendations to customers now becomes possible. Collaborative recommender systems utilize correlation between customer preference ratings to identify "like-minded" customers and predict their product preference. One factor determining the success of the recommender systems is the prediction accuracy, which in many cases is limited by lacking adequate ratings (the sparsity problem). Recently, the use of latent class model (LCM) has been proposed to alleviate this problem. In this paper, we first study how the LCM can be extended to handle customers and products outside the training set. In addition, we propose the use of a pair of LCMs (called dual latent class model-DLCM), instead of a single LCM, to model customers' likes and dislikes separately for enhancing the prediction accuracy. Experimental results based on the EachMovie dataset show that DLCM outperforms both LCM and the conventional correlation-based method when the available ratings are sparse.
KW - Collaborative fitering
KW - Latent class models (LCMs)
KW - Personalization
KW - Recommender systems
UR - http://www.scopus.com/inward/record.url?scp=0742307431&partnerID=8YFLogxK
U2 - 10.1109/TSMCA.2003.818877
DO - 10.1109/TSMCA.2003.818877
M3 - Journal article
AN - SCOPUS:0742307431
SN - 1083-4427
VL - 34
SP - 143
EP - 148
JO - IEEE Transactions on Systems, Man, and Cybernetics Part A:Systems and Humans
JF - IEEE Transactions on Systems, Man, and Cybernetics Part A:Systems and Humans
IS - 1
ER -