三维微阵列数据的多目标进化聚类

Multi-Objective Evolutionary Triclustering of 3D Microarray Data

聚类技术广泛应用于微阵列数据分析中。在基因-样本-时间GST微阵列数据矩阵中,挖掘三维聚类成为当前的热门研究课题。3D聚类过程经常需要对多个相互冲突的目标进行优化,而且进化算法以其强大的探寻能力成为高维搜索空间中非常有效的搜索方法。本文基于多目标进化计算方法提出一个新的3D聚类算法MOE-TC,以挖掘GST数据中的3D聚类。现实微阵列数据上的实验验证结果充分说明了本文算法的有效性。

The clustering technique is widely used in microarray data analysis, and mining three-dimensional（3D） clusters in gene-sample-time（simply GST） microarray data is emerging as a hot research topic in this area. During the mining of 3D clusters, several objectives have to be optimized simultaneously, and often these objectives are in conflict with each other. Moreover, with great exploration power, evolutionary computation is made as an effective search approach in the search space of huge dimensionality. Based on MOEA（Multi-Objective Evolutionary Algorithm）, this paper proposes a new 3D cluster algorithm, MOE-TC（Multi-Objective Evolutionary TriClustering）, tO mine 3D clusters in 3D mieroarray data. Experimental results in real microarray data confirm the validity of the proposed technique.