Social Web Recommendation using Metapaths.
Robin Burke, Fatemeh Vahedian · 2013
The social web is characterized by a wide variety of connections between individuals and entities. A challenge for social web recommendation is make the most effective use of a diverse set of relations. Typically, researchers focus on a limited set of relations (for example, person to person ties for user recommendation or annotations in social tagging recommendation). In this paper, we propose a general approach to recommendation in social networks that can incorporate multiple relations in combination. A key feature of this approach is the use of the metapath, an abstraction of a large class of paths in the network in which edges of different types are traversed in a particular order. As a preliminary demonstration, we show that our approach yields improvements over a state-of-the-art technique on several social tagging datasets. General Terms Hybrid recommender system, heterogeneous network, social tagging system