基于不同数据类型的隐结构模型在中医研究中的应用与进展＊

隐结构模型是一种特殊的聚类分析方法,可对临床收集的无监督数据进行多维聚类划分,能直观地显示出数据间的关系、揭示数据间的隐藏规律,近年来在中医研究中应用广泛,其研究结果为中医临床辨证论治提供了客观的依据。但因其分析过程复杂,对样本数据要求严谨,在一定程度上限制了其在中医研究中的应用。本文从既往的研究结果入手,基于不同数据类型对隐结构模型在中医研究中的成果进行分类总结,着重论述了隐结构模型对不同类型数据的研究内容和适用性,以及隐结构模型联合其他数据分析方法的应用,并对其中存在的问题进行总结与思考,为隐结构模型对不同类型数据的分析应用以及中医量化研究提供思路和借鉴。

Application Progress of Latent Structure Model Based on Different Data Types in Traditional Chinese Medicine Research

Latent structure model is a special clustering analysis method, which can carry out multi-dimensional clustering division on the unsupervised data collected in clinic. It can intuitively show the relationship between the data and reveal the hidden law between the data. In recent years, it has been widely used in the research of TCM. Its research results provide an objective basis for the clinical syndrome differentiation and treatment of TCM. However, due to its complex analysis process and strict requirements for sample data, its application in TCM research is limited to a certain extent. Starting with the previous research results, this paper classified and summarized the achievements of implicit structure model in TCM research based on different data types, focused on the research content and applicability of latent structure model to different types of data, as well as the application of latent structure model combined with other data analysis methods. It summarized and considered the existing problems, and provided ideas and reference for the analysis and application of implicit structure model to different types of data and quantitative research of traditional Chinese medicine.