A learning based self-organized additive fuzzy clustering method and its application for EEG data

Issue title: Intelligent Information Processing: Techniques and Applications

Guest editors: C.P. Limx and L.C. Jainy

Article type: Research Article

Authors: Kuwata, Tomoyuki^a | Sato-Ilic, Mika^a | Jain, Lakhmi C.^{b; *}

Affiliations: [a] Faculty of Systems and Information Engineering, University of Tsukuba, Tsukuba, Ibaraki, Japan | [b] School of Electrical and Information Engineering, University of South Australia, Adelaide, South Australia, SA, Australia | [x] University of South Australia, Adelaide, South Australia, Australia | [y] University of Science, Malaysia

Correspondence: [*] Corresponding author. E-mail: [email protected]

Abstract: In this paper, a learning based fuzzy clustering method and its application to a set of electroencephalogram (EEG) data is given. The proposed method combines the learning process of noise to a conventional self-organized additive fuzzy clustering method. This is done by using the inner product of a pair of degrees of belongingness of objects. By learning the status of the noise in each iteration of the algorithm, the proposed method can obtain a more adaptable result.

Keywords: Fuzzy clustering, learning noise, self-organized similarity, electroencephalogram (EEG) data

DOI: 10.3233/KES-2012-0238

Journal: International Journal of Knowledge-based and Intelligent Engineering Systems, vol. 16, no. 1, pp. 69-78, 2012