定向判别分析新算法及应用  

A New Algorithm for Directional Discriminant Analysis and Its Application

作　　者：丁跃潮[1] 万春[1] 孙扬[1] 

机构地区：[1]集美大学计算机工程学院,福建厦门361021

出　　处：《计算机工程与科学》2006年第9期77-79,93,共4页Computer Engineering & Science

基　　金：福建省自然科学基金资助项目(A0410021);福建省教育厅科技项目(JA004235)

摘　　要：本文介绍了多元有序数据定向判别分析新方法的原理、建模流程、应用流程和应用实例。这种判别分析将分类建模与判别归类分开。新方法用多组或逐步判别分析对多元有序数据建模,应用时根据应用领域的知识对样本归属作初步定向,然后选择模型的相关局部进行判别归类。这种方法解决了由于时间序列多元数据周期性造成的样本分类颠倒问题。This paper introduces the principle, modeling flowchart, application flowchart and a practical example of a new algorithm called Directional Discriminant Analysis （DDA）, which may be used in multivariate sequence data. In DDA, class modeling and data discriminating are separated. The model for multivariate sequence data is built by multiple or stepwise discriminant analysis. In applying the model, the initial estimation of the samples＇ classification should be given according to the knowledge in the application field. Then the program selects the appropriate part of the model to discriminate the classes of the data. In this way, we can solve the upside down problem of sample classification caused by the periodicity of multivariate time series data.

关 键 词：判别分析 多元数据 定向判别 建模 最优分割 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构] TP311[自动化与计算机技术—计算机科学与技术]