NEIGHBORHOOD-BASED APPROACH OF COLLABORATIVE FILTERING TECHNIQUES FOR BOOK RECOMMENDATION SYSTEM

Recommendation System or Recommender System help the user to predict the "rating" or "preference" a user would give to an item. Recommender systems in general helps the users to find content, products, or services (such as digital products, books, music, movie, TV programs, and web sites) by combining and analyzing suggestions from other users, which mean rating from various people, and users. These recommendation systems use analytic technology to calculate the results that a user is willing to purchase, and the users will receive recommendations to a product of their interest. The aim of the System is to provide a recommendation based on users likes or reviews or ratings. Recommendation system comprises of content based and collaborative based filtering techniques. In this paper, collaborative based filtering has been used to get the expected outcome. The expected outcome has been achieved through collaborative filtering with the help of correlation techniques which in turn comprises of Pearson correlation, cosine similarity, Kendall’ s Tau correlation, Jaccard similarity, Spearman Rank Correlation, Mean-squared distance, etc. This paper tells about which similarity metrics such us Pearson correlation (PC), constrained Pearson correlation (CPC), spearman rank correlation (SRC) which is good in the context of book recommendation system and then applied with neighborhood algorithm