Abstract: It proposes quantum-behaved particle swarm optimization QPSO algorithm on the basis of the PSO algorithm and applies it to a data set on gene expression.The proposed clustering algorithm partitions the [N] patterns of the gene expression dataset into user-defined K categories to minimize the fitness function of total within-cluster variation.Based on the merits of K-means algorithm and using K-means clustering to seed the initial swarm，combing QPSO，PSO to cluster data，it proposes KQPSO，KPSO algorithm.The experiment results on four gene expression data sets using K-means，PSO，QPSO，KPSO，KQPSO five clustering algorithms show that the QPSO-based clustering algorithm has a good performance in gene expression data analysis.

Key words: gene expression data, clustering, Quantum-behaved Particle Swarm Optimization（QPSO） algorithm

摘要： 在PSO算法的基础上提出的基于量子行为的QPSO算法，并将其应用到基因表达数据集上。QPSO基因聚类算法是将N条基因根据使TWCV（Total Within-Cluster Variation）函数值达到最小分到由用户指定的K个聚类中。根据K-means算法的优点，利用K-means聚类的结果重新初始化粒子群，结合QPSO和PSO的聚类算法提出了KQPSO和KPSO算法。通过在4个实验数据集上利用K-means、PSO、QPSO、KPSO、KQPSO 5个聚类算法得出的结果比较显示QPSO算法在基因表达数据分析上具有良好的性能。

关键词: 基因表达数据, 聚类, 基于量子行为的粒子群优化（QPSO）算法