Accurate Grid-based Clustering Algorithm with Diagonal Grid Searching and Merging

2017 3rd International Conference on Applied Materials and Manufacturing Technology

Liu, Feng^1 ; Ye, Chengcheng^1 ; Zhu, Erzhou^1

School of Computer Science and Technology, Anhui University, Hefei

230601, China^1

关键词: Data analysis methods;    Data mining technology;    Data space;    Density parameters;    Grid-based clustering algorithm;    Vertical down;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/242/1/012123/pdf
DOI  :  10.1088/1757-899X/242/1/012123

来源: IOP

【 摘 要 】

Due to the advent of big data, data mining technology has attracted more and more attentions. As an important data analysis method, grid clustering algorithm is fast but with relatively lower accuracy. This paper presents an improved clustering algorithm combined with grid and density parameters. The algorithm first divides the data space into the valid meshes and invalid meshes through grid parameters. Secondly, from the starting point located at the first point of the diagonal of the grids, the algorithm takes the direction of "horizontal right, vertical down" to merge the valid meshes. Furthermore, by the boundary grid processing, the invalid grids are searched and merged when the adjacent left, above, and diagonal-direction grids are all the valid ones. By doing this, the accuracy of clustering is improved. The experimental results have shown that the proposed algorithm is accuracy and relatively faster when compared with some popularly used algorithms.