Design of computationally efficient density-based clustering algorithms

作者：

Highlights：

• Proposed a new strategy to reduce computational complexity associated with the DBSCAN

• Developed new density based algorithms based on correlation measure

• Cluster analysis on two synthetic and six real datasets demonstrates the performance of proposed method.

• An interesting application is demonstrated to identify the regional hazard regions present in seismic catalog of Japan.

摘要

•Proposed a new strategy to reduce computational complexity associated with the DBSCAN•Developed new density based algorithms based on correlation measure•Cluster analysis on two synthetic and six real datasets demonstrates the performance of proposed method.•An interesting application is demonstrated to identify the regional hazard regions present in seismic catalog of Japan.

论文关键词：Clustering, classification, and association rules,Mining methods and algorithms,DBSCAN,Fast DBC,Physical action datasets,Seismic catalog of Japan

论文评审过程：Received 1 September 2012, Revised 5 May 2014, Accepted 24 November 2014, Available online 29 November 2014.

论文官网地址：https://doi.org/10.1016/j.datak.2014.11.004