Feature Selection Based Approach for Handling Cold Start Problem in Collaborative Recommender Systems

Due to the widespread usage of Information and communication technology (ICT), nowadays people are getting a large number of options to choose a particular item or a service. So, in this scenario, the recommender system (RS) plays a very vital role to optimize their decisions. But cold start problem is one of the major challenges in RS for new users and items. In this paper, a novel method using feature selection and prediction has been proposed to address the cold start problem in Collaborative RS. The proposed approach has been validated using two data sets i.e Laptop Dataset and Red wine Quality dataset taking Mean Absolute Error (MAE) and Precision as the evaluation metrics.