A Scrutable Algorithm for Enhancing the Efficiency of Recommender Systems using Fuzzy Decision Tree

Abstract

Recommender system plays the major role of filtering the needed information from enormous amount of overloaded information. From e-commerce to movie websites, recommender systems are being used for market their product to the customer. Also, recommender system gains user trust by suggesting the customer's products of interest based on the profile of the customer and other related information. So, when the recommender system goes wrong or suggests an irrelevant product, the customer will stop trusting and using the recommender system. This kind of scenario will affect the customer as well as the e-commerce and other websites that depends on recommender systems for boosting the sales. There is a significant need to correct the recommender system when it goes wrong, since, wrong recommendations will weaken the user trust and diminish the efficiency of the system. In this paper, we are defining a scrutable algorithm for enhancing the efficiency of recommender system based on fuzzy decision tree. Scrutable algorithm will correct the system and will work on enhancing the efficiency of the recommender system. By adapting the scrutable algorithm, users will be in a position to understand the transparency in recommending items which, in turn, will gain user trust.