Title :
GMDBSCAN: Multi-Density DBSCAN Cluster Based on Grid
Author :
Xiaoyun, Chen ; Yufang, Min ; Yan, Zhao ; Ping, Wang
Author_Institution :
Sch. of Inf. Sci. & Eng., Lanzhou Univ., Lanzhou
Abstract :
DBSCAN is one of the most popular algorithms for cluster analysis. It can discover all clusters with arbitrary shape and separate noises. But this algorithm canpsilat choose parameter according to distributing of dataset. It simply uses the global MinPts parameter, so that the clustering result of multi-density database is inaccurate. In addition, when it is used to cluster large databases, it will cost too much time. For these problems, we propose GMDBSCAN algorithm which is based on spatial index and grid technique. An experimental evaluation shows that GMDBSCAN is effective and efficient.
Keywords :
data mining; grid computing; pattern clustering; very large databases; GMDBSCAN; cluster analysis; global MinPts parameter; grid technique; large databases; multidensity DBSCAN cluster; multidensity database; spatial index; Algorithm design and analysis; Clustering algorithms; Costs; Information analysis; Information science; Multi-stage noise shaping; Partial response channels; Shape; Spatial databases; Spatial indexes; Clustering; Data Mining; Local_MinPts; Multi-Density; SP-Tree; Unit Grid Density;
Conference_Titel :
e-Business Engineering, 2008. ICEBE '08. IEEE International Conference on
Conference_Location :
Xi´an
Print_ISBN :
978-0-7695-3395-7
DOI :
10.1109/ICEBE.2008.54
