基于伪F统计FAMC算法的基因表达数据分析

Gene Expression Data Analysis of FAMC Algorithm Based on Pseudo F-statistics

基因芯片技术在给人类带来巨大机遇的同时也带来一些挑战。针对基因表达数据的海量性,以及基因类属的不确定性等问题,提出了一种基于伪F统计量(PFS)的模糊属性均值聚类FAMC(fuzzy attribute c-means cluste-ring)算法,并就模糊参数m的确定问题提出了有效的解决方法。最后将其在标准的基因表达数据上进行测试分析,取得了较优的聚类结果。

Gene chip technology not only brings the huge opportunity to the humanity but also brings some challenges simultaneously. In view of gene expression data＇s magnanimous, uncertainty about the gene class and so on, this article proposed one kind fuzzy attribute c-means clustering algorithm based on Pseudo F-statistics, and proposed an effective solution for defining the fuzzy parameter m. Finally carried on it in the standard gene expression data for testing analysis, obtained superior cluster results.