有不少群聚方法被用來分析LR-型模糊數。然而，這些方法有不同程度的缺點。為了克服這些聚類分析法的問題，在本文中，提出了一種利用自我組織程序的新群集演算法來處理LR-型模糊數。使用這個群集方法，LR-型模糊數可以利用彼此距離的相似度而自動群集，得到好的分群結果。
此外，本饗所提出的群集方法，也可以從模糊數中識別出極端值。為了檢查此方法的有效性，我們還應用學生的成績表現及病人的血壓等二個真實資料的例子，結果顯示，對於這些真實資料，分群的結果也是好的。

Several cluster methods were used for analysis of LR-type Fuzzy data. However, those measures suffered different levels of drawbacks. For overcoming the existing problems in those clustering methods, in this paper, we proposed a new clustering method based on a self-organization procedure for handling LR-type fuzzy numbers. When the proposed clustering algorithm was employed, the LR-type data could be self-organized by using their distance of similarity and resulted in a good clustering classification. In addition, the proposed clustering method could find out the outlier from the Fuzzy data set. For examining the effectiveness of the proposed approach, we then apply this algorithm to two real data sets which are students’ learning performance and patients’ blood pressure data. The results indicated that the proposed method did obtain good clustering results for these real data sets.

[1] L.A. Zadeh, Fuzzy sets, Inf. Control 8 (1965) 338-353.
[2] S.F. Schnatter, On statistical inference for fuzzy data with applications to descriptive statistics, Fuzzy Sets and Systems 50(1992) 143-165.
[3] M.S. Yang, C.H. Ko, On a class of fuzzy c-numbers clustering procedures for fuzzy data, Fuzzy Sets and Systems 84(1998) 49-60.
[4] W.L. Hung, M.S. Yang, Fuzzy clustering on LR-type fuzzy numbers with an application in Taiwanese tea evaluation, Fuzzy Sets and Systems 150(2005) 561-577.
[5] P. D’Urso, P. Giordani, A weighted fuzzy c-means clustering model for fuzzy data, Comput. Stat. Data Anal. 50(2006) 1496-1523.
[6] T.L. Chen, S.Y. Shiu, A new clustering algorithm based on self-updating process. Proceedings of the American Statistical Association. (2007).
[7] W.L. Hung, M.S. Yang and E.S. Lee, 2010, A robust clustering procedure for fuzzy data, Computers and Mathematics with Applications, vol. 60, no. 1, pp. 151-165.
[8] T. Denoeux, M.H. Masson, Principal component analysis of fuzzy data using autoassociative neural networks, IEEE Trans. Fuzzy Syst. 12 (2004) 336-349.