The document discusses content recommendation in adaptive social networks based on data mining. It aims to design a methodology for social recommender systems that incorporate different knowledge sources from structured and unstructured data. The objectives are to design improved explanations for recommendations to increase user acceptance and enhance the student experience. The approach uses a hybrid recommender system that adapts the weighting of collaborative and content-based filtering based on the type of content being recommended. Current results show the system integrated into a Brazilian social network with over 70,000 students and items, with early user feedback being positive. Expected results include analyzing how recommendations can improve the learning process and exploring hidden knowledge in social networks.