User correlation model for question recommendation in community question answering

In this paper, we address the problem of question recommendation that automatically recommends a new question to suitable users to answer in community question answering (CQA). The major challenge of question recommendation is the accurate selection of suitable users to answer. Most of the existing approaches attempt to find suitable users in CQA by estimating user’s existing capability, user’s interest or blending both for the question. However, these methods ignore correlation among users (askers and answerers) in terms of topic preference. In this study, we propose a user correlation model (UCM) to effectively estimate degree of correlation among users in terms of topic preference. Furthermore, we present the UCM-based approach to question recommendation, which provides a mechanism to naturally integrate the correlation between answerer and asker in terms of topic preference with content relevance between the answerer and the question into a unified probabilistic framework. Experiments using real-world data from Stack Overflow show that our UCM-based approach consistently and significantly improves the performance of question recommendation. Hence, our approach can increase question recommendation accuracy in CQA according to utilize the correlation between answerer and asker in terms of topic preference.