A Review Paper on Recommender System Type’s and Evaluation Metrices

Abstract

A very simple definition of Recommender Systerm is “Any system that produces personalized
recommendations as output or helps the user in guiding in a personalized way to get interesting or useful objects
from a large collection of useful options.”This Review paper presents the study in the field of recommender systems
and describes the current generation of recommendation methods w h i c h i s usually divided into the
three main categories: content-based recommender system, collaborative recommender system,Utility Based
,Knowledge Based and hybrid recommendation techniques.This paper prescribes different terms applicable to an
even broader range of applications for improvement in accuracy of recommendation algorithms. The research carried
out has focused on improving the accuracy of recommender systems. In this paper, we propose that the
recommender system should move beyond the conventional accuracy criteria and take some other criteria into
account, such as coverage, diversity, serendipity, scalability, adaptability, risk, novelty and so on. These extensions
include, among others, improvement of understanding of users and items, incorporation of the contextual
information into the recommendation process, support for multi-criteria ratings, and provision of more flexible and
less intrusive types of recommendations.