中文摘要：

为了进一步提高双聚类结果的性能,提出了一种基于变分贝叶斯的半监督双聚类算法。首先,在双聚类过程中引入了行和列的辅助信息,并提出了相应的联合分布概率模型;然后基于变分贝叶斯学习方法对联合概率分布中的参数进行估计;最后,通过合成数据集和真实的基因表达式数据集对提出的算法性能进行评估。实验表明,提出的算法在进行双聚类分析时,其归一化互信息量明显优于相关的双聚类算法。

英文摘要：

In order to improve the performance of result by biclustering algorithms, this paper proposed a variational Bayes based semi-supervised bielustering algorithm. Firstly, it introduced the row and column auxiliary information, and proposed a corresponding joint distribution probahilistic model. Secondly, it estimated the parameters of the joint prohabilistic distribution based on variational Bayes learning. Finally, it validated the performance of the proposed algorithm with synthetic and real gene expression datasets. The experiments show that, while evaluating the performance of biclustering algorithms, the normalized mutual information of the proposed algorithm is obviously higher than related works.