Intelligent question answering system based on Artificial Neural Network

The ability of the machine to infer knowledge from the user documents can be tested based on its ability to answer the question asked. Conventional Artificial Neural Network (ANN) models for knowledge extraction only answer to the questions which are simple and objective as they don't analyze the questions and don't try to understand what really the content of document mean. The proposed question answering system (QAS) uses deep cases along with ANN to understand the contents of the documents. We divide the sentences of natural language into knowledge units and assign deep case to each word to improve the quality of knowledge extraction.