Published Paper Details:

Social recommendation systems are critical in directing users to appropriate material, goods, and relationships in the age of digital connectedness. This review offers a thorough examination of two fundamental elements that have a considerable effect on the performance of these systems: homophily and influence. Homophily, or the tendency for people to associate with others who have similar features or interests, and influence, or the ability of some entities to impact the behavior of others, both have significant implications for the design and optimization of social recommendation algorithms. This study starts by delving into the theoretical underpinnings of homophily and influence, providing insights into their psychological bases and societal significance. It then discusses how these phenomena present themselves in various online communities, such as social networking sites, marketplaces, and discussion boards. This article reviews 25 research papers for homophily and influence for social recommendation and explores the potential of computer-assisted methods for social recommendation and staging. A significant amount of this analysis is devoted to chronicling the many computational models and approaches used in social recommendation systems to measure and utilize homophily and influence. These models span from classic matrix factorization to cutting-edge deep learning architectures. We investigate the advantages and disadvantages of each strategy, giving light on their application to various recommendation areas.