Collaborative Filtering Based Personalized Hybrid Recommendation System Using Machine Learning Techniques

Abstract

The sizzling growth of e-commerce platforms, online social networking websites, and online media has led to abundant information and choices, making it challenging for users to find what they want. Recommendation systems have arisen to solve this problem, proposing customized and relevant things to users based on their previous behaviour, interests, and context. Collaborative Filtering (CF) is an important part of recommendation systems approaches and it is most widely used in implementation of personalized recommendation systems. The machine learning techniques such as Matrix Factorization (MF) is a popular technique used in collaborative filtering, which is used to extract underlying factors from the user-item rating matrix. Another technique such as Neural Network (NN) has also been widely used in recommendation systems. In this paper, the use of matrix factorization with neural network have incorporated and suggested.