A Hindi Question Answering System using Machine Learning approach

A Question Answering (QA) System is fairly an Information Retrieval(IR) system in which a query is stated to the system and it relocates the correct or closest results to the specific question asked in natural language. It is one of the consequences of Natural Language Interface to Database (NLIDB). The paper discusses the implementation of a Hindi Language QA system developed using Machine Learning approach. The implemented QA system is divided into three phases: Accessing natural language (NL) Query; where the input query is read, preprocessed and get tokenized; next is feature extraction (FE) phase; where specific features vectors are identified from the results of previous phase and finally the Classification phase; where the Naïve Baye's classifier has been used, along with the knowledge base already stored in the system. This paper reflects that the concepts of similarity and classification provide better results than the use of `equals' concept by defining the overall accuracy of finding the relevant answers of the specific questions asked by the user.