Many computational techniques have been proposed by social networks to analyze the users' behaviors to recommend relevant content for them. Social networks generate a huge volume of information, which users cannot consume, generating a problem known as information overload. This way, filtering relevant information to help users with this problem becomes necessary. Social networks have many available features, such as relationships and interactions, which can be used to investigate the users' behaviors regarding news on their feed. The value of news can be defined as Social Capital, which is used by this work to model the user's preferences. This paper aims to investigate, model, and quantify interactions on social networks by exploiting social capital to develop a recommender system. Hence, in order to evaluate recommendations, an experiment was conducted with real users. Results show that our proposal was able to generate relevant recommendations on at least 62\% of the scenarios.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.