A Novel Clustering Data based on K-Means

Abstract

In this paper a new algorithm for clustering symbolic data based on K-Means algorithm is proposed .This new algorithm allows the data entry and the membership degree to be intervals. In our approach, we propose a dynamic document clustering based on structured MARDL technique. In this method, each document is assigned a weight by term frequency and inverse document frequency method using cosine similarity measure and then, the documents are first separated into clusters using k-Means method. The largest cluster will split and forms two sub clusters and this step would be repeated for many times until clusters formed are with high similarity. In addition, our approach tends to capture the intrinsic structure of a data set, e.g., the number of clusters. Simulation results demonstrate that our approach yields favorite results for a variety of temporal data clustering tasks. As our weighted cluster ensemble algorithm can combine any input partitions to generate a clustering ensemble, we also investigate its limitation by formal analysis and empirical studies.