Contextual recommender system for E-commerce applications

Abstract

Today's arena of global village organizations, social applications, and commercial websites provides huge information about products, individuals, and activities. This is leading to a plethora of content that requires effective handling to obtain the desired information. A recommendation system (RS) suggests relevant items to the user according to his/her desired preference. It processes various information related to users and items. However, RSs suffer from data sparsity. Generally, deep learning techniques are used in RSs for deep analysis of item contents to create precise recommendations. However, the effective handling of user reviews in parallel with item reviews is still an open research domain that can be further explored. In this paper, a hybrid model that handles both user and item metadata concurrently is proposed with the aim of solving the sparsity problem. To demonstrate the viability of the proposed methodology, a series of experiments was performed on three real-world datasets. The results show that the proposed model outperforms other state-of-the-art approaches to the best of our knowledge. (C) 2021 Elsevier B.V. All rights reserved.