Survey on Recommendation Systems

Abstract

In recent decade's recommendation systems (RSs) plays an essential role in many applications as World Wide Web. Also recommendation system is one of the most important research area in machine learning. Recommendation system functions as a helper to find the interest of users by making relevant suggestions to users. The RSs mainly use four filtering methods to provide personalized recommendations to users, the most popular ones are: Collaborative filtering (CF), Content-based filtering, Demographic filtering and hybrid filtering. Data mining is one of the important analysis techniques used in RSs to predict user interest in information, products and services among the vast amount of available items. The data mining techniques that are most commonly used in RSs are: classification, clustering and association rule discovery. This paper performs a survey on recommendation systems, techniques, challenges and issues and lists some research papers solve these obstacles, also data mining methods used in recommender systems.