Recommendation system is becoming an important part of electronic commerce systems. And collaborative filtering is the most hot research topic for building electronic commerce personalized recommendation system and is extensively used in many fields. Collaborative filtering aims at predicting a target user's ratings for new items by integrating other like-minded users' rating information. The user-based approach is a common technique used in collaborative filtering. This method first uses statistical approaches to measure user similarities based on their previous ratings on different items. Users will then be grouped into different neighborhood according to the calculated similarities. Finally, the approach will generate predictions on how a user would rate a specific item by aggregating ratings on the item cast by the identified neighbors of the target user. Collaborative filtering algorithm usually suffers from two fundamental problems: sparsity and scalability. In this paper, the problems of sparsity and scalability are described. And an overview of collaborative filtering recommendation algorithm in electronic commerce is presented. © 2012 Springer-Verlag GmbH.
CITATION STYLE
Huang, Y. (2012). Research of collaborative filtering recommendation algorithm in electronic commerce. In Advances in Intelligent and Soft Computing (Vol. 159 AISC, pp. 615–620). https://doi.org/10.1007/978-3-642-29387-0_95
Mendeley helps you to discover research relevant for your work.