Community question answering (CQA) has become a very popular web service to provide a platform for people to share knowledge. In current CQA services, askers post their questions to the system and wait for answerers to answer them passively. This procedure leads to several drawbacks. Since new questions are presented to all users in the system, the askers can not expect some experts to answer their questions. Meanwhile, answerers have to visit many questions and then pick out only a small part of them to answer. To overcome those drawbacks, a probabilistic framework is proposed to predict best answerers for new questions. By tracking answerers’ answering history, interests of answerers are modeled with the mixture of the Language Model and the Latent Dirichlet Allocation model. User activity and authority information is also taken into consideration. Experimental results show the proposed method can effectively push new questions to the best answerers.
Weitere Kapitel dieses Buchs durch Wischen aufrufen
Bitte loggen Sie sich ein, um Zugang zu diesem Inhalt zu erhalten
Sie möchten Zugang zu diesem Inhalt erhalten? Dann informieren Sie sich jetzt über unsere Produkte:
- Predicting Best Answerers for New Questions in Community Question Answering
- Springer Berlin Heidelberg
Neuer Inhalt/© ITandMEDIA