Tiered Document Clustering Through Finest Centroid Assortment

Abstract

Data growth is enormously higher today and difficult to handle in an organized manner and in effective time limit. In data mining, a wide range of research is available for managing data effectively. Similarly, document clustering is one of the notable firm process in the data mining utilized to assemble the related documents. In this article, we generate an algorithm for clustering by means of Adaptive Pillar K-means and Gaussian Firefly Algorithm (GFA). For determining the proper centroid in order to attain the proper clustered documents, Adaptive Pillar K-means Algorithm is utilized. Subsequently, GFA is exploited for the optimization process and also for enhancing the precision that results in reducing the sum of squared errors and computational time. Here, the performance of the proposed methodology is compared with various algorithms such as Genetic Algorithm, Ant colony optimization, and gravity clustering. The attained results shows the performance of the proposed methodology and the simulation results illustrated the betterment in quality with low sum of squared errors.