Text Document Clustering Based on Neural K-Mean Clustering Technique

Abstract

Data clustering is a significant tool for applications like search engines and document browsers. It gives the user an overall vision of the information contained in the data sets. The well-known techniques of data clustering do not look for exact problems of clustering like high dimensionality of the dataset, large size of the datasets and to understand the ability of the cluster description. The work done before does not have the inbuilt property of clustering, so, here for extracting the features from the document, the clusters of different classes present in the document are taken. In the proposed work, this problem can be solved by using similarity technique on neighbors in addition to K- means method.