Comparative Analysis of Functionality and Aspects for Hybrid Recommender Systems

Abstract

Recommender systems are gradually becoming the backbone of profitable business which interact with users mainly on the web stack. These systems are privileged to have large amounts of user interaction data used to improve them. The systems utilize machine learning and data mining techniques to determine products and features to suggest different users correctly. This is an essential function since offering the right product at the right time might result in increased revenue. This paper gives focus on the importance of different kinds of hybrid recommenders. First, by explaining the various types of recommenders in use, then showing the need for hybrid systems and the multiple kinds before giving a comparative analysis of each of these. Keeping in mind that content-based, as well as collaborative filtering systems, are widely used, research is comparatively done with a keen interest on how this measures up to hybrid recommender systems.