Question answering system provides the way which helps us in reducing the time of search for the useful information from huge amount of data. An extensive work has been done in the field of Question Answering systems but there exists a scope of further improvement in this field. In the proposed architecture the answers are extracted by correctly classifying the questions. Paragraph ranking is used to reduce the text that reduces the memory as well as processing requirement. Named Entity Recognition technique helps to increase the accuracy of the answer returned. The proposed implementation takes question as input from user, classifies the question and then attempts to find answer that will be based on corresponding answer type. The techniques like paragraph ranking, preprocessing, indexing etc are used to improve the efficiency and accuracy of the system. Thus the system provides accurate and relevant answer of the query without making so much effort and also helps in reducing overall time in searching for answers. Moreover the result returned will be very concise.

Indexing Named Entity Recognition Question Classification Ranking Summarization