Interval-Valued Centroids in K-Means Algorithms

The K-Means algorithms are fundamental in machine learning and data mining. In this study, we investigate interval-valued rather than commonly used point-valued centroids in the K-Means algorithm. Using a proposed interval peak method to select initial interval centroids, we have obtained overall quality improvement of clusters on a set of test problems in the Fundamental Clustering Problem Suite (FCPS).

[1]  Lihi Zelnik-Manor,et al.  Approximate Nearest Subspace Search , 2011, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[2]  J. Mendel,et al.  Fast Computation of Centroids for Constant-Width Interval-Valued Fuzzy Sets , 2006, NAFIPS 2006 - 2006 Annual Meeting of the North American Fuzzy Information Processing Society.

[3]  Greg Hamerly,et al.  Alternatives to the k-means algorithm that find better clusterings , 2002, CIKM '02.

[4]  M. P. S Bhatia,et al.  Data clustering with modified K-means algorithm , 2011, 2011 International Conference on Recent Trends in Information Technology (ICRTIT).

[5]  Vladik Kreinovich,et al.  Knowledge Processing with Interval and Soft Computing , 2008 .

[6]  Doheon Lee,et al.  Fuzzy clustering of categorical data using fuzzy centroids , 2004, Pattern Recognit. Lett..

[7]  Padhraic Smyth,et al.  From Data Mining to Knowledge Discovery in Databases , 1996, AI Mag..

[8]  Ramasamy Uthurusamy,et al.  Data mining and knowledge discovery in databases , 1996, CACM.