Much of my research is in Topic Modeling. I currently work on Community Question Answering and try to propose Topic Modeling solutions for problems in that community.
One of those problems refers to finding the relevant answers for a new question using existing answers in the forum structured webstites.
But my main focus is on expert finding in CQA. I model the StackOverFlow users interests by tracking users history . I am also working on Language Model approaches to compare the results of Topic Modeling approach with it. In the following you can find some related works to what I am doing now:
- Predicting Best Answerers for New Questions in Community Question Answering, by:Qing Yang et al.
- The Author-Recipient-Topic Model for Topic and Role Discovery in Social Networks, with Application to Enron and Academic Email, by: Andrew Mcllum et al.
- Expertise Modeling for Matching Papers with Reviewers, by: Andrew Mcllum et al.
- Topics over Time: A Non-Markov Continuous-Time Model of Topical Trends, by Andrew Mcllum et al.
- The Author-Topic Model for Authors and Documents, by: Michal Rosen-Zvi et al.
- Dynamic Topic Models, by: David Blie et al.
- Latent Dirichlet Allocation, by: David Blie et al.
- Routing Questions to the Right Users in Online Communities, by:G. Cong et al.
- Braod Expertise Retrieval in Sparse Data Environments, by: Krisztian Balog et al.
- Formal Models for Expert Finding in Enterprise Corpora, by: Krisztian Balog et al.