Knowledge Based Deep Learning Collaborative Filtering (KBDLCF)

Abstract

Since from past few years huge volume of users are interacting with website for their day to day transactions are being going up. In this regards numbers of companies are encroaching towards internet to tell their products and services. This revolutions towards E-Commerce has changed, Conventional way of doing business this rapid expansion has resulted in new challenges to both companies as well as customers. Thus customers are overloaded with multiple choices for individual product which results in a confused and lost state. It has become a trivial for the webmasters to evaluate whether products and services are provided are catering the needs of customer or not. Recommender system has been playing a more vital and essential role in various information access systems to boost business and facilitate decision-making process.in this paper we proposed Knowledge Based Deep Learning Collaborative Filtering has been presented. The use of this approach joining the advantages of fuzzy logic and neural networks in order to improve CF recommender system that recommends items to the active user on the basis of fuzzy rules. The execution of neural networks for collaborative filtering recommendations was enlightened by an example real world dataset of movies and Reviews rated by user i.e Netflix. Results of a simulation study using various similarity measures and recommendation methods show that the KBDLCF model performs better than other conventional techniques. The experimental results reveal that forming ambiguity using fuzzy logic and neural networks mends the performance of personalized recommender systems.