Recommender Systems have become an important part of large e-commerce websites. One can safely say, they are the bread and butter of large E-Commerce websites. We may have seen the "customers who bought item1 also bought item2" feature of sites such as amazon. com and found it useful. This is exactly what recommender systems strive to achieve. The basic notion behind introducing recommender systems in websites is simple: to help the customers or users using the website in making their decisions. In general the goal of any recommendation system is to present users with a highly relevant set of items. Recommendation algorithms can be generally classified into three types (i) Non-Personalized, (ii) Content-Based, and (iii) Collaborative Filtering algorithms. Apart from these three approaches, we also have hybrid approach wherein we can combine the above mentioned algorithms to improve the performance of recommender systems. Literature survey done on recommender systems shows that a lot of work has been carried out in this area and this paper gives an insight into two very popular recommender systems: Non-personalized and Collaborative recommender systems. The paper gives an insight into two approaches of Non-personalized recommender systems and the User-based approach of Collaborative recommender systems.

User-Based Collaborative Filtering Technique Item-Based Collaborative Filtering Content-Based Filtering Pearson-Correlation.