Neural network: an exploration in document retrieval system

As more and more information is stored electronically, the demand for intelligent methods becomes increasingly urgent. The basic belief underlying this research is that a truly helpful document retrieval system must “understand” what the user is looking for. A neural network has information processing structure attributes for adaptation to fulfil the needs within an information environment. These attributes are suitable to be used in a document retrieval system in order to build a faster, efficient and user-friendly system. This paper presents experimental research on the effectiveness of a neural network model in document retrieval. The main purpose of this research is to demonstrate the feasibility of the proposed approach. The backpropagation neural network learning method is used to build up and employ application domain knowledge for the document retrieval system. The knowledge is acquired from examples of queries and relevant documents. Then, another collection of queries is used to test the effectiveness of the system