A New Clustering Method for Minimum Classification Error

Heo Gyeongyong; Seong Hoon Kim

Journal of The Korea Society of Computer and Information

Abbr : JKSCI
2014, 19(7), pp.1-8
Publisher : The Korean Society Of Computer And Information
Research Area : Computer Science

Heo Gyeongyong ¹ , Seong Hoon Kim ²

¹ 동의대학교
² 경북대학교

ABSTRACT

Clustering is one of the most popular unsupervised learning methods, which is widely used toform clusters with homogeneous data. Clustering was used to extract contexts corresponding toclusters and a classification method was applied to each context or cluster individually. However,it is difficult to say that the unsupervised clustering is the best context forming method from theview of classification. In this paper, a new clustering method considering classification was proposed. The proposedmethod tries to minimize classification error in each cluster when a classification method is applied to each context locally. For this purpose, the proposed method adds constraints forcing two datapoints belong to the same class to have small distances, and two data points belong to differentclasses to have large distances in each cluster like in linear discriminant analysis. The usefulnessof the proposed method is confirmed by experimental results.
KEYWORDS

Clustering , Supervised clustering , Context , Classification
REFERENCES (13)

[journal] L. A. Zadeh / 1965 / Fuzzy sets / Information and Control 8 (3) : 338~353

[journal] E. H. Ruspini / 1969 / A new approach to clustering / Information and Control 15 (1) : 22~32

[journal] Rui Xu / 2005 / Survey of Clustering Algorithms / IEEE Transactions on Neural Networks 16 (3) : 645~678

[journal] J. C. Dunn / 1974 / A fuzzy relative of the ISODATA process and its use in detecting compact well separated clusters / Journal of Cybernetics 3 (3) : 32~57

[book] J. C. Bezdek / 1981 / Pattern Recognition with Fuzzy Objective Function Algorithms / Plenum

[confproc] G. Heo / 2010 / A Noise Robust Variant of Context Extraction for Local Fusion / Proceedings of 2010 IEEE International Conference on Fuzzy Systems : 1~6

[journal] A. M. Bensaid / 1996 / Partially supervised clustering for image segmentation / Pattern Recognition 29 (5) : 859~871

[journal] A. M. Martinez / 2001 / PCA versus LDA / IEEE Transactions on Pattern Analysis and Machine Intelligence 23 (2) : 228~233

[confproc] D. E. Gustafson / 1978 / Fuzzy clustering with a fuzzy covariance matrix / Proceedings of the 1978 IEEE Conference on Decision and Control : 761~766

[confproc] R. Babuska / 2002 / Improved Covariance Estimation for Gustafson-Kessel Clustering / Proceedings of the 2002 IEEE International Conference on Fuzzy Systems : 1081~1085

[journal] B. Feil / 2007 / Geodesic Distance Based Fuzzy Clustering / Advances in Soft Computing 39 : 50~59

[journal] Gomes, Ryan / 2010 / Discriminative clustering by regularized information maximization / Advances in Neural Information Processing Systems 23 : 775~783

[journal] K. Zhang / 2009 / Maximum Margin Clustering Made Practical / IEEE Transactions on Neural Networks 20 (4) : 583~596
KCI Citation (1)

KJC Korea
Journal Central

Statistics

Tools