An Evaluation of the Objective Clustering Inductive Technology Effectiveness Implemented Using Density-Based and Agglomerative Hierarchical Clustering Algorithms

摘要

The paper presents the results of the research concerning comparison analysis of the efectiveness of OPTICS and DBSCAN density-based and agglonarative hierarchical clustering algorithms within the framework of the objective clustering inductive technology. Implementation of this technology allows us to determine the optimal parameters of appropriate clustering algorithm in terms of the maximum values of the complex balance criterion which contains as the components both the internal and the external clustering quality criteria. The data from the Computing School of East Finland University database were used as the experimental one during the simulation process. The results of the simulation have shown high effectiveness of the proposed technique. The investigated objects were divided into clusters correctly in all cases. Moreover, the results of the simulation have shown also higher effectiveness of the density-based clustering algorithms in comparison with agglomerative hierarchical algorithm use due to more level of the detail during the objects clustering.