A directional clustering technique for random data classification

This paper introduces a new clustering technique for random data classification based on an enhanced version of the Voronoi diagram. This technique is optimized to deal in the best way possible with data distributions which in their spatial representations experience overlap. A mathematical framework is given in view of this enhanced analysis and provides insight to key issues involving (a) the use of a correction process to complement the traditional Voronoi diagram and (b) the introduction of directional vectors in Gaussian and elliptical data distributions for enhanced data clustering. The computational requirements of the proposed approach are provided, and the computer results involving both randomly generated and real-world data prove the soundness of this clustering technique.