Social context analysis for topic-specific expert finding in online learning communities

Abstract

With increasing trend toward lifelong learning, online learning communities have become important places for people to seek and share knowledge. Yet with the increasing number of members and produced artifacts within the learning communities, it is challenging to find the related documents and influential experts who post topic-specific high-quality content. In the context of Web2.0, Web documents, e.g., blogs or forum messages, are freely discussed and commented by users. The users-generated content (e.g., comments) and activities (e.g., forward) implicitly reflect the importance of Web documents. Therefore, the social context, integrating the document content information and the social event information, are analyzed in this chapter to discover expertise in online learning communities. In addition to computing documents’ topic-focus degree, the proposed approach measures the quality of documents according to users’ feedback behaviors, review sentiment, and topic-specific influence of users who give feedback. Experiments on real dataset have shown that our approach is effective to find the meaningful topic-specific expertise. In online learning communities, this approach could be used to identify and recommend experts with high expertise and influence to make community interconnected and cohesive.