基于多核学习的电子病历多视图聚类研究及应用

Research and application of multi-view clustering of electronic health records based on multi-kernel learning

电子病历的分析和应用是智慧医疗领域研究的热点,提出一种基于多核学习的电子病历多视图聚类模型,对多核学习挖掘电子病历的非线性相关信息和高阶相关性信息、利用谱聚类对融合的核矩阵(相似性矩阵)进行聚类研究,得到多视图数据空间的聚类结果。实验结果表明,该方法能快速准确地对电子病历中的信息进行聚类,在性能指标上优于其它方法。

The analysis and application of electronic medical records is a hot topic in the field of smart healthcare.A multi view clustering model for electronic medical records based on multi kernel learning is proposed.The model utilizes multi kernel learning to mine nonlinear and high-order correlation information of electronic medical records,and spectral clustering to fuse the kernel matrix(similarity matrix)for clustering research.The clustering results of the multi view data space are obtained.The experimental results show that this method can quickly and accurately cluster information in electronic medical records,and is superior to other methods in terms of performance indicators.