Web document classification using fuzzy K-nearest neighbor

Abstract

With surge in the number of documents across the internet, increasing the efficiency of any retrieval model is a challenging task. As non-relevant information is retrieved across the internet, increasing the accuracy of any search model is one of the research concerns. Fuzzy classification is broadly applied to address the search issue in search engines. Fuzzy logic provides a methodology to interpret natural language using membership functions. A variant of k-Nearest Neighbor (kNN) called Fuzzy kNN is explored in this paper. This paper provides a comparative analysis of results obtained using kNN and Fuzzy kNN. The Fuzzy kNN results obtained show significant improvement.