CRLRM: Category Based Recommendation Using Linear Regression Model

Abstract

A system that suggests list of most popular items to a set of users on the basis of their interest is named as recommendation system. Recommendation system filters the unnecessary information by applying knowledge discovery techniques for online users and has become the most powerful and admired tools in E-Business. ERPM is one of the easiest movie recommendation method, which overcomes the limitations of scalability and sparsity of recommendation system, but it generates predictions on the basis of probability model, which are less accurate and requires more time for calculations. This article presents a novel method named CRLRM (Category based Recommendation using Linear Regression Model) which is based on linear regression model that improves the prediction accuracy and speed up the calculations. Performance of proposed method is evaluated on the basis of MAE (Mean Absolute Error) comparison, and result obtained is far much better than ERPM and shows improvement in 30-40% of user ratings.