Resumen
In this paper, a new method has been proposed to eliminate the weaknesses in the previous algorithms. The proposed method for data density clustering is reduced in the mapping programming model. Our analysis result shows that misleading data was presented to prove the function of the density-based clustering algorithm and the weakness of the base method on them has been represented. Then, local clustering was tested by competing methods for standard data clustering and its superiority to these methods was determined. When passing local clustering to distributed clustering, misleading data was again used to prove the quality of clustering. Distributed clustering quality is lower than local clustering, but it is still superior to the base method. The quality of clustering of the proposed method on competing methods was clearly determined by distributed network clustering. Finally, the method of choosing this parameter was described by evaluating the homogeneity and completeness criteria and the effect of the flexible parameter on different types of data.