A Fuzzy Weighted Density-Based Clustering Algorithm for Industrial Databases

摘要：

Cluster analysis is an important work of data mining, and the current algorithms are not suitable for the industrial database, which is generally a large, highdimension and coupling database. The paper proposes a fuzzy weighted density-based clustering algorithm for industrial databases. The proposed algorithm, which is based on DBSCAN, uses the fuzzy weighted distance between objects to instead of the general distance between objects of DBSCAN. In the clustering process,the weights of dimensions are not fixed all alone.Especially, the weights of coupling dimensions are adjusted according to the fuzzy logic rules. We performed the experiments by using a test dataset and a real industrial database. The experiments results verify the proposed algorithm not only can do clusters analysis on industrial database successfully but also has the higher clustering effectiveness.

关键词： Cluster analysis industrial database fuzzy weighted distance density-based clustering algorithm

作者: Hui Cao Gangquan Si Yanbin Zhang Lixin Jia

作者单位: The Electrical Engineering School Xian Jiaotong University Xian, Shaanxi, 710049, P.R.China

会议类型: 国际会议

会议名称: The International Colloquium on Onformation Fusion 2007(2007年国际信息融合研讨会)

会议地点: 西安

会议语种:英文

页码: 404-410

在线出版日期: 2007-08-22（万方平台首次上网日期，不代表论文的发表时间）