Hybrid Cohort Rating Prediction Technique to leverage Recommender System

Abstract

The long tail of diverse consumption of resources online by the customers raises a challenge for the e-commerce websites and service providers. Recommender system offers a vigorous way to cope up with the aforementioned challenge. In this paper, we have proposed a hybrid cohort rating prediction technique which relies on high cohort users and high cohort items to make predictions. Our model significantly improves the retention of recommender system showing encouraging results when compared with existing traditional recommender systems.