Product Recommender System Using Collaborative Filtering Technique

Abstract

There is a rapid increase of data on internet, to enhance the search result as required and give the better results based on their personalization, we introduce Item Of Interest (IOI). Recommender System (RS) is an engine used to predict the future interest of set of items for user and suggest top N items. New methods in Recommender System are required to evolve change in finding the Item Of Interest (IOI) and reduce data sparsity problem. Recommender Systems are used in many streams such as product sales, recommending websites, books, movies, etc. The three types of RS are Collaborative Filtering (CF), Content based Filtering (CBF), and Hybrid Filtering (HF). In this work we are experimenting Item Based Collaborative Filtering. Item Based Collaborative Filtering is a technique where we acquire information from item and process them to predict the item of interest (IOI). We used three clustering technique to form clusters among products. Amazon dataset is used to experiment our work. Finally, we evaluated the experimental result using Root Mean Square Error (RMSE) and Mean Absolute Error (MAE).