Double fuzzy C-means model and its application in the technology innovation of China

Issue title: Some highlights on fuzzy systems and data mining

Guest editors: Shilei Sun, Silviu Ionita, Eva Volná, Andrey Gavrilov and Feng Liu

Article type: Research Article

Authors: Li, Li^a | Wang, Renxiang^a | Li, Xican^{b; *}

Affiliations: [a] School of Economics, Wuhan University of Technology, Wuhan, P.R. China | [b] School of Information Science and Engineering, Shandong Agricultural University, Taian, P.R. China

Correspondence: [*] Corresponding author. Xican Li, School of Information Science and Engineering, Shandong Agricultural University, Taian, P.R. China. Tel.: +86 138 5389 7651; Fax: +86 053 8824 9275; E-mail: lxc@sdau.edu.cn.

Abstract: Aiming to improve the algorithm of the classic fuzzy C-means model (FCM), a double fuzzy C-means model (DFCM) was presented in this paper. A new fuzzy cluster validity index (RWW) and the DFCM algorithm were proposed, simultaneously. Then, the double fuzzy C-means model was applied for the clustering analysis of the regional technology innovation level in China. The validity of the double fuzzy C-means model was tested using the wine data set of UCI. The comparison results of different cluster validity indexes validated the fuzzy cluster validity index (RWW) proposed in this paper. The application example and wine data set clustering results indicated that the DFCM model enhanced the intra-class compactness and inter-class separation, making the classification more accurate.

Keywords: Double fuzzy C-means model, category membership, cluster validity index, technology innovation level

DOI: 10.3233/JIFS-169172

Journal: Journal of Intelligent & Fuzzy Systems, vol. 31, no. 6, pp. 2895-2901, 2016