基于聚类与主成分分析的HRV时域指标敏感性研究

BASED ON CLUSTER ANALYSIS PRINCIPAL COMPONENT ANALYSIS OF TIME DOMAIN SENSITIVITY INDEX OF HRV RESEARCH

指标敏感性是多数客观实验研究都要面临的问题,多数研究中的敏感性主要是指标的灵敏度分析,而对于各项研究的指标群内各指标对所研究问题的贡献程度即指标重要性方面的敏感性研究还比较缺乏。本文以客观实验数据为基础,采用了聚类分析和主成分分析相结合的方法,通过指标群内各组合客观实验数据的个体贡献率的大小确定了短时程HRV时域指标的敏感性,给今后HRV指标选择敏感性研究提供一种可行的参考方法和理论依据。

Index sensitivity was most objective experimental study are facing problems,most research is mainly the sensitivity analysis of the sensitivity of index,and for the research of the index of each index in the study of the problems contribution degree index importance to the sensitivity of the aspects that the relative lack of research.Based on test data,this paper detennines the sensitivity of short- term HRV time domain indexes,by using principal component analysis and cluster analysis method,through each combination objective experimental data individual contribution rate in index clusters.And the results provide a feasible reference methods and theoretical basis for future HRV index selection sensitivity research.