期刊
ADVANCES IN DATA ANALYSIS AND CLASSIFICATION
卷 2, 期 3, 页码 211-225出版社
SPRINGER HEIDELBERG
DOI: 10.1007/s11634-008-0032-5
关键词
Multi-objective optimization; Clustering; Fuzzy C-means; Microarray; Gene expression data
资金
- Centre National de la Recherche Scientifique
- Institut National de la Sante et de la Recherche Medicale
- Universite Louis Pasteur de Strasbourg
In this paper, we use the Fuzzy C-means method for clustering 3-way gene expression data via optimization of multiple objectives. A reformulation of the total clustering criterion is used to obtain an expression which has fewer variables compared to the classical FCM criterion. This transformation allows the use of a direct global optimizer in constrast to the alternating search commonly used. Gene expression data from microarray technology is generally of high dimension. The problem of empty space is known for this kind of data. We propose in this paper a transformation allowing more contrast in distances between all pairs of data samples. This, hence, increases the likelihood of detecting group structure, if any, in high dimensional datasets.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据