RANKING RESULTS IN A WEB BASED COMMUNITY QA SYSTEM

Abstract

Web forums like Stack overflow, Yahoo Answer and Quora comes under community question answering system
(CQA) are gaining popularity. CQA archives of previously asked questions and their answers, this means that one can
freely ask any question on web forums and expect some good, honest answers. On other side, it takes effort to go through
all possible answers of search question and to make sense of them but this process take too much time like popular
question to have hundreds of answers, and it is very time-consuming for a user to inspect them all. To resolve this
problem thus system mainly works on ranking of QA pairs and online forum. This QA pairs are ranked by fine grained
and QA forwarding technique. This technique ranks based on rating given by common user. This technique explicitly
captures relationships between questions and their best answers by modeling topical dependencies. In some of condition
unsatisfied answers on system resolved by online forum system with direct communication. In QA system not only user
can post question and answer, but also it can correct spell at the time of posting question and answer, this system used
fuzzy spell check API