K-means是一個大家所熟知的演算法。Ahmed et al. 在2002年提出調整鄰近點的強健方法。Chen and Zhang在2004年使用中位數代替平均數改進Ahmed et al.的方法。Huh and Lim在2009年提出變異權重的K-means演算法。本篇我們提出一個強健的特徵權重K-means演算法，從一些模擬數值和真實數據的實驗和多變異權重K-means演算法做比較，我們發現我們提出的演算法可以獲得較好的實驗結果。

Fuzzy c-means clustering (FCM) with spatial constraints (FCM_S) is an effective algorithm suitable for image K-means are the most well-known conventional clustering methods. Ahmed et al. (2002) proposed to increase the robustness of the neighborhood terms by directly modifying the objective function Algorithms. Chen and Zhang (2004) Use the median instead of mean improvement of Ahmed et al.'s Method. Huh and Lim(2009) proposed a weighting variables in K-means clustering algorithms. In this paper, we propose a Feature-weight selection in k-means algorithms with correction. Some numerical and real case experiments are performed to assess the performance of weighting variables in K-means clustering in comparison with Feature-weight selection in k-means algorithms with correction. Experimental results show that the proposed a Feature-weight selection in k-means algorithms with correction has better performance.

6. References
[1] Huh, Myung-Hoe and Lim, Yong B, 2009. Weighting variables in K-means
clustering, Journal of Applied Statistics.
[2] Hung, W.L.,Yang, M.S., Chen, D.H., 2008. Bootstrapping approach to feature-weight
selection in fuzzy c-means algorithms with an application in color image gmentation.
29, 1317-1325.
[3] J.Z. Huang, M.K. Ng, H. Rong, and Z. Li, 2005. Automated variable weighting in k-means type clustering. IEEE Trans. Pattern Anal. Machine Intelligence 27, 657-668.
[4] Ahmed, M.N., Yamany, S.M., Mohamed, N., Farag, A.A., Moriarty, T., 2002. A
modified fuzzy c-means algorithm for bias field estimation and segmentation of MRI
data. IEEE Trans. Med. Imaging 21, 193-199.
[5] Chen, S.C., Zhang, D.Q., 2004. Robust image segmentation using FCM with spatial
constrains based on new kernel-induced distance measure. IEEE Trans. Systems Man
Cybernet. Pt. B 34, 1907–1916.