Evaluation of Similarity Functions by using User based Collaborative Filtering approach in Recommendation Systems
Shaivya Kaushik, Pradeep Tomar · International Journal of Engineering Trends and Technology · 2015
Recommendation Systems has been comprehensively analysed and are changing from novelties used by a few E-commerce sites in the past decades.Many of the popular and largest commerce websites are widely using recommendation systems.These are popular and important part of the e- commerce ecosystem that help users to find relevant and valuable information through large product spaces.The tremendous growth of visitors and the information poses few key challenges such as producing high quality recommendation systems,performing many recommendation systems per second for millions of users and items.The paper introduce user based collaborative filtering approach and the similarity function.The algorithm will identify relationships between different users and then compute recommendation for the users.This paperpresent a most commonly used similarity functions and their computation that aims to determine which similarity function result in producing most accurate recommendation.