A NEURO-FUZZY BASED DOCUMENT TRACKING & CLASSIFICATION SYSTEM

Abstract

The worldwide web is an extraordinary resource for gaining access to information of all kinds. Each day a greater number of resources become available online. The advantages that internet offers researchers are tremendous; so much that some may be tempted to bypass the library entirely and conduct all their research on the web. In spite of this, the enquirers are being confronted with information overloaded with noise which the conventional search methods lack the capability to tackle. Existing search engines such as Google, Yahoo and Bing often return a long list of results which forces users to sift through it to find relevant documents; thereby making search for information difficult. This paper takes a critical study of document tracking and classification systems and presents a Neuro-fuzzy based model for classification of search results based on the strength of words in the query. The structured systems analysis and design methodology and the neuro-fuzzy clustering technique were employed to study and classify the documents into groups of similar topics for specific knowledge. The work built an adaptive intelligent information retrieval system which will cluster internet documents into similar topic using an unsupervised machine learning techniques to reduce the percentage of irrelevant documents that are retrieved and presented to users.