Document Clustering Using Divisive Hierarchical Bisecting Min Max Clustering Algorithm

Abstract

Document clustering is a process of grouping data object having similar properties. Bisecting kmeans is a top down clustering approach wherein all the documents are considered as single cluster. That cluster is then partitioned into two sub-clusters using k-means clustering algorithm, so k is considered as 2. Sum of square errors (SSE) of both the clusters are calculated. The cluster which has SSE greater, that cluster is split. This process is repeated until the desired number of clusters are obtained. Divisive Hierarchical Bisecting Min–Max Clustering Algorithm is similar to bisecting k-means clustering algorithm with a slight modification. To obtain a certain number of clusters. The main cluster is divided into two clusters using Min-Max algorithm. A cluster is selected in order to split it furthers. This process is repeated until the desired number of clusters are obtained. Divisive Hierarchical Bisecting Min–Max Clustering Algorithm is similar to bisecting k-means clustering algorithm with a slight modification. To obtain a certain number of clusters. The main cluster is divided into two clusters using Min-Max algorithm. A cluster is selected in order to split it furthers. This process is repeated until desired numbers of clusters are obtained.