Recommender systems have been developed to alleviate information overload, aid user decision-making, and achieve different forms of personalization. Their effectiveness and usability have been demonstrated in a number of applications, including e-commerce, movies, music, travel, and social networks. The same information filtering and personalization needs are now arising in the area of educational experiences and resources. The educational learning environment is no longer limited to in-class lectures, both teaching and learning can be taken place on the Web. For example, Technology Enhanced Learning (TEL) and Massive Open Online Courses (MOOC) are two of the most popular applications that can benefit from the application of recommendation technology. Other popular applications related to recommenders in the educational domain include book recommendations for school-aged readers (i.e., K-12) as well as and the recommendation of informal learning programs.

A variety of recommendation techniques can be used to assist educational recommendations, such as semantic or content-based recommender systems, transfer learning, or collaborative intelligence. Traditional strategies, however, are not sufficient in within the academic environment, as the generated suggestions are based on needs and expectations beyond user/content similarity/historical data. The availability of more heterogeneous information (such as friendships, fellowships, social media, interactions across multiple devices, user behaviors on multiple categories of items or activities) increases the demand to (i) effectively leverage these information sources to learn how they can interact in identifying suitable items to recommend and influence users’ preferences in the educational recommender systems, and (ii) exploit these information to better suggest appropriate items (e.g., books, courses, programs, degrees, activities) to the end users.