Incorporating User Item Similarity in Hybrid Neighborhood-Based Recommendation Systems

Recommendation systems have been developed in many domains to help users with information overload from the large volume of online multimedia content by providing them appropriate options. Recently hybrid recommendation require a amount data understand users’ interests and give suggestions. However, several internet privacy issues make skeptical about sharing their personal service providers, limiting potential these systems. The study this paper introduces various novel methods utilizing baseline estimate learn user specific item features past interactions. Subsequently, extracted feature vectors are implemented user-item correlations, an additional fine-tuning factor for neighborhood-based collaborative filtering Comprehensive experiments show that similarity scores rating prediction task can improve accuracy at least 2.11% compared traditional while minimizing need tracking users' digital footprints.