TY - JOUR
T1 - Interest before liking
T2 - Two-step recommendation approaches
AU - Zhao, Xiangyu
AU - Niu, Zhendong
AU - Chen, Wei
PY - 2013/8
Y1 - 2013/8
N2 - Recommender systems become increasingly significant in solving the information explosion problem. Existing techniques focus on minimizing predicted rating errors and recommend items with high predicted values to people. They consider high and low rating values as liking and disliking, respectively, and tend to recommend items that users will like in the future. However, especially in the information overloaded age, we consider whether a user rates an item as a measure of interest no matter whether the value is high or low, and the rating values themselves represent the attitude to the quality of the target item. In this paper, we propose two-step recommendation approaches that recommend items matching users' interests first, and then try to find high quality items that users will like. Experiments using MovieLens dataset are carried out to evaluate the proposed methods with precision, recall, NDCG, preference-ratio and precision-like as evaluation metrics. The results show that our proposed approaches outperform the seven existing ones, i.e., UserCF, ItemCF, ALS-WR, Slope-one, SVD++, iExpand and LICF.
AB - Recommender systems become increasingly significant in solving the information explosion problem. Existing techniques focus on minimizing predicted rating errors and recommend items with high predicted values to people. They consider high and low rating values as liking and disliking, respectively, and tend to recommend items that users will like in the future. However, especially in the information overloaded age, we consider whether a user rates an item as a measure of interest no matter whether the value is high or low, and the rating values themselves represent the attitude to the quality of the target item. In this paper, we propose two-step recommendation approaches that recommend items matching users' interests first, and then try to find high quality items that users will like. Experiments using MovieLens dataset are carried out to evaluate the proposed methods with precision, recall, NDCG, preference-ratio and precision-like as evaluation metrics. The results show that our proposed approaches outperform the seven existing ones, i.e., UserCF, ItemCF, ALS-WR, Slope-one, SVD++, iExpand and LICF.
KW - Binary user model
KW - Collaborative filtering
KW - Recommender system
KW - Two-step recommendation
KW - User interests
UR - http://www.scopus.com/inward/record.url?scp=84878436885&partnerID=8YFLogxK
U2 - 10.1016/j.knosys.2013.04.009
DO - 10.1016/j.knosys.2013.04.009
M3 - Article
AN - SCOPUS:84878436885
SN - 0950-7051
VL - 48
SP - 46
EP - 56
JO - Knowledge-Based Systems
JF - Knowledge-Based Systems
ER -