Latent Structure Analysis and Syndrome Differentiation for Integration of Traditional Chinese Medicine and Western Medicine (II): Joint Clustering
隐结构分析与西医疾病的辨证分型（Ⅱ）：综合聚类

In an earlier paper, we explained how in principle latent structure analysis can provide objective evidence for the syndrome differentiation of patients diagnosed as suffering from a western medicine disease. Indeed, latent structure analysis can systematically reveal various facets of symptom data collected from those patients and yield a partition of the data along each of those facets. In this paper, we point out the need to perform further cluster analysis of the data by jointly considering information from different facets and propose a method for the task. The method is applied to a heart disease data set of 3021 cases of cardiovascular diseases. A number of partitions with clear syndrome interpretations have been obtained.