Using lower-bound similarity to enhance the performance of recommender systems

Authors:
Mohak Sharma;P. Krishna Reddy
Affiliations:
International Institute of Information Technology, Hyderabad, India;International Institute of Information Technology, Hyderabad, India
Venue:
COMPUTE '11 Proceedings of the Fourth Annual ACM Bangalore Conference
Year:
2011

Citing 7
Cited 1

GroupLens: an open architecture for collaborative filtering of netnews

CSCW '94 Proceedings of the 1994 ACM conference on Computer supported cooperative work
GroupLens: applying collaborative filtering to Usenet news

Communications of the ACM
Analysis of recommendation algorithms for e-commerce

Proceedings of the 2nd ACM conference on Electronic commerce
An Empirical Analysis of Design Choices in Neighborhood-Based Collaborative Filtering Algorithms

Information Retrieval
A collaborative filtering algorithm and evaluation metric that accurately model the user experience

Proceedings of the 27th annual international ACM SIGIR conference on Research and development in information retrieval
Toward the Next Generation of Recommender Systems: A Survey of the State-of-the-Art and Possible Extensions

IEEE Transactions on Knowledge and Data Engineering
Clustering with Lower Bound on Similarity

PAKDD '09 Proceedings of the 13th Pacific-Asia Conference on Advances in Knowledge Discovery and Data Mining

Improving the performance of recommender system by exploiting the categories of products

DNIS'11 Proceedings of the 7th international conference on Databases in Networked Information Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

Recommender systems employ the popular K-nearest neighbor collaborative filtering (K-CF) methodology and its variations for recommending the products. In K-CF approach, recommendation for a given user is computed based on the ratings of K-nearest neighbors. In K-CF approach, it can be noted that, the system identifies K neighbors for each user irrespective of the number of products he/she has rated. As a result, the user who have rated few products may get the less-similar neighbors and the user who have rated more products may miss the genuine neighbors. In the literature, the notion of lower-bound similarity has been proposed to improve the clustering performance in which the clusters are extracted by fixing the similarity threshold. In this paper, we have extended the notion of lower-bound similarity to recommender systems to improve the performance of K-CF approach. In the proposed approach, instead of fixing K for finding the neighborhood, the similarity threshold value is fixed to extract the neighbors for each user. As a result, each user gets appropriate number of neighbors based on the number of products rated by him/her in a dynamic manner. The experimental results, on MovieLens dataset, show that the proposed lower bound similarity CF approach improves the performance of recommender systems over K-CF approach.