摘要
FCM算法在基因表达数据分析中存在噪声点,影响聚类结果,为此提出了一种改进的模糊核聚类算法,通过使用Mercer核把原始数据映射到高维特征空间,并为特征空间的每个向量分配一个动态权值,分析权值的大小来识别噪声点,得到一个较为理想的聚类结果.实验结果表明,该方法比FCM聚类算法具有更好的聚类效果.
FCM algorithm has been widely applied in gene expression analysis. It＇s effectiveness and efficiency in analysing gene expression data, however, is somewhat limited since it is sensitive to noise and since gene expression data often contain noise. Therefore, a method of improved fuzzy kernel clustering algorithm is presented in this paper. Through mercer kernel functions, the data in the original space are mapped to a high-dimensional feature space, and an additional weighting factor is assigned to each vector in the feature space. The noise could find through analysis weight of a data, then, dusters can be identified. The experiment results show that the proposed method is more effective than FCM algorithm.
出处
《江南大学学报（自然科学版）》
CAS
2006年第2期162-165,170,共5页
Joural of Jiangnan University (Natural Science Edition)
关键词
模糊聚类
核函数
特征空间
生物信息学
fuzzy clustering
kernel function
feature space
bioinformatics
