An Approach for Document Clustering using Agglomerative Clustering and Hebbian-type Neural Network

Abstract

Clustering is a useful method that categorizes a large quantity of unordered text documents into a small number of meaningful and coherent collections, thereby providing a basis for instinctive and informative navigation and browsing mechanisms. Different type of distance functions and similarity measures have been used for clustering, such as squared, cosine similarity, Euclidean distance and relative entropy. This paper presents text document space dimension reduction in text document retrieval by agglomerative clustering and Hebbian-type neural network. Hebbian-type neural network reduce document space to two dimensions so each document is represented as a point in the reduced document space. Furthermore, the clusters are formed in compact document space.