A Geometry-Based Accelerated Fusion Clustering Algorithm and its Application in Marine Engineering

• DOI: 10.4028/www.scientific.net/amr.108-111.106
• Publication Date: 2010-05-04T19:09:08Z
• Authors: Tian Zhen Wang; Yang Liu; Tian Hao Tang
• Keywords: 01 natural sciences; 0105 earth and related environmental sciences

In order to solve the problem in k-means algorithm that inappropriate selection of initial clustering centers often causes local optimum and time complexity is too high when handling large amounts data, a fusion based on geometry proposed this paper. The result experiments shows better than traditional k-means++ algorithms, with higher quality faster speed. And at last paper, we apply it marine engineering.