Development of Deep Learning-based Media Content Recommendation System, DL-MCRS, for User Satisfaction

Abstract

The ever-growing quantity of audio-visual information accessible today may be effectively managed by recommender systems, which assist users in discovering new and interesting topics. An increasing number of customized suggestion apps have emerged on the World Wide Web in the last decade. Recommendation systems can’t function without precise behaviour modelling of users. The conventional wisdom about friend suggestion algorithms leaves out crucial user data, leading to a misleading portrayal of their actions. The common understanding of friend suggestions is inaccurate because it disregards crucial user information. Hence, this paper proposes that the Deep Learning-Based Media Content Recommendation System (DL-MCRS) improves efficiency and user satisfaction by integrating huge multi-source heterogeneity data and building more precise user and item models on social media platforms. The suggested method uses the semantic personalized recommendation system (SPRS) to bridge the gap between high-level semantic information and low-level media properties. The suggested system uses domain ontology to customise video recommendations to their interests based on a user’s past actions on the site. The experimental findings show that the suggested strategy outperforms the baseline methods concerning efficiency.