Analysis of User-based and Item-based Prediction Algorithms for Recommender System

Abstract

Recommender Systems typically use techniques from collaborative filtering which recommend items that users with similar preferences have liked in the past and, also predict new rating by averaging ratings between pairs of similar users or items. Predictions come from three sources: predictions based on ratings of the same item by other users, predictions based on different item ratings made by the same user, and ratings predicted based on data from other but similar users rating other but similar items. In this system, we use prediction algorithms to provide users with items that match their interests based on collaborative filtering (CF) approach. Our system use similarity measures between users, and also between items from a single rating criteria .We provide analysis of user-based and item-based prediction algorithms. The accuracy of the algorithms is compared by Mean Absolute Error.