Social networks have become an indispensable part of our lives, which serve as communication channels, social interaction platforms as well as ubiquitous entertainment tools; meanwhile, social networks constantly generate big social media data that create decision complexity and information overload to users. As a result, recommender systems are emerged to suggest personalized and possibly preferred media for the users. However, social networks have extensively enriched the inputs for recommender systems, such as users' social relations, data source credibility, and new social media types. Consequently, this paper is aimed at identifying the crucial factors that can be used to advance recommender systems in social networks. For each factor, this paper discusses the state-of-the-art recommender system research in that aspect, and suggests how to integrate the featured data to build and improve recommender systems for social networks. The paper further proposes a model to integrate the crucial factors and indicates possible application domains for social media recommender systems.
Advanced recommender systems by exploiting social networks
D'Auria D
2019-01-01
Abstract
Social networks have become an indispensable part of our lives, which serve as communication channels, social interaction platforms as well as ubiquitous entertainment tools; meanwhile, social networks constantly generate big social media data that create decision complexity and information overload to users. As a result, recommender systems are emerged to suggest personalized and possibly preferred media for the users. However, social networks have extensively enriched the inputs for recommender systems, such as users' social relations, data source credibility, and new social media types. Consequently, this paper is aimed at identifying the crucial factors that can be used to advance recommender systems in social networks. For each factor, this paper discusses the state-of-the-art recommender system research in that aspect, and suggests how to integrate the featured data to build and improve recommender systems for social networks. The paper further proposes a model to integrate the crucial factors and indicates possible application domains for social media recommender systems.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.