A study on different closed domain question answering approaches

Question answering (QA) framework is a framework that gives answers to the inquiries raised by the client using the common language. The framework recovers minor portion of the content from the collection of the report, which contains the appropriate response for the client's inquiry. In order to retrieve such response from the repository, information retrieval techniques are needed and for further processing or comprehension of the client's inquiry, presented in the characteristic language, natural language processing techniques are utilized. However to make the recovering procedure increasingly hearty, snappy and accurate, the idea of knowledge-based classification also included in this work, for this reason, utmost care was taken in training the framework. using "Jaccard likeness", the closest answer for the client's inquiry was reached. In addition to this, "WordNet" was used to recover the appropriate response, depends on both syntactic and semantic similitudes. Utilizing these ideas we have actualized a QA framework on space "Hyderabad Tourism" which gives in general exactness of 92%. In this work, our main aim is to create a closed-domain question answering framework, which will give the precise and considerably short answer to all the inquiries that are related to the Hyderabad city, as a response, instead of giving a lengthy paragraph or document.