User-Based Collaborative Filtering. Based on Improved Similarity Algorithm

摘要

With the fast development of World Wide Wed,Web-based applications and services should allow user to getthe right personalized information quickly and effectively.Collaborative Filtering acts a very important role in webservice personalization and Recommender System. In thispaper, Stability Degree was proposed to improve the accuracyof User based collaboration filtering, three kinds of StabilityDegree were introduced into similarity computation, and theresults show that the prediction accuracy can be improved byII percents, and MAE can be reduced faster than classicmethod.