An Efficient Community Question Answer (CQA) using the Metadata Rating

Community Question Answering (CQA) is the question answering process. Question Answering (QA) is the field of information retrieval and Natural language Processing (NLP). In this paper we mainly focus on lexical gap and word embedding problem. Here, the CQA aims to find the existing question which is equivalent to queried question. The other existing system of this paper is two novel category powered models. That is basic category powered model called as MB-NET and enhanced category powered model called as ME-NET. They are used to learn about the lexical gap and word embedding problem. The proposed system of this paper is to extend the metadata information easily using user ratings, like signals and Poll and Survey signals, into the learning process to obtain more powerful word representations.