Incorporating Interest Preference and Social Proximity into Collaborative Filtering for Folk Recommendation

In many social communities, it is increasingly popular for people to seek useful information or resources from reliable peers (i.e., folks). In this regard, folk recommendation is no less important than other types of recommendation such as book recommendation, movie advertisement, etc. In this paper, we focus on incorporating user similarity (in terms of interest similarity and social proximity) with user-based collaborative filtering (CF) for folk recommendation. Specifically, we target at recommending folks (i.e. new trusted users, or friends) to a given user in an existing social community network. To this end, a range of similarity-based and CF-based algorithms are evaluated by using two real-world application datasets, demonstrating their potential for effective and efficient folk recommendation.