计算机与现代化

• 算法设计与分析 • 上一篇    下一篇

基于AUC的非参数快速变点检测算法

  

  1. 广东工业大学自动化学院,广东广州510006
  • 收稿日期:2015-03-12 出版日期:2015-07-23 发布日期:2015-07-28
  • 作者简介:吴学龙(1990-),男,山东临沂人,广东工业大学自动化学院硕士研究生,研究方向:模式识别; 徐维超(1970-),男,山东青岛人,教授,研究方向:模式识别与智能系统。
  • 基金资助:
    国家自然科学基金资助项目(61271380); 广东省自然科学基金资助项目(S2012010009870,1414050001981)

A Nonparametric Fast Algorithm for Abrupt Changes Detection Based on AUC

  1. School of Automation, Guangdong University of Technology, Guangzhou 510006, China
  • Received:2015-03-12 Online:2015-07-23 Published:2015-07-28

摘要: 目前,变点检测技术已经广泛应用到各个领域。然而,由于实际生产环境的复杂性,变点检测技术中的常用参数方法往往存在一定的局限性。为了克服这些问题,提出一种新的变点检测的非参数方法,通过互联网检索应为首次应用AUC(曲线下面积)对样本数据进行在线变点检测。该方法将变点检测分为2个阶段:预分析阶段,对样本数据进行加窗处理,通过计算窗口中样本数据的AUC值的方式来间接得到其均值和方差;检测阶段,通过假设检验的方法对经过处理后的样本数据进行变点检测。通过实验仿真,可以观察到此算法与常规CUSUM算法相比,具有更好的稳健性,而且对检测多个突变点的情况同样有效。

关键词: 变点检测, 非参数估计, AUC

Abstract: Recently, the technologies for detecting abrupt changes have been prevalent in many fields. However, as the conventional strategy, parametric method has some inherent limitations when it is used in real production environment. In order to solve these problems, we propose a novel nonparametric method for abrupt changes detection, which solves the problem by AUC (Area under the Curve). We divided the process into two phases: during the preanalysis period, we set two continuous scanning windows and then calculate the AUC values of the samples in difficult scanning windows in order to obtain the means and variances; during the detection period, we detect the changes by hypothesis testing. In simulation experiments, the method we proposed outperforms CUSUM in terms of robustness and multiple changes detection.

Key words: abrupt changes detection, nonparametric method, area under the curve(AUC)

Key words: abrupt changes detection,
nonparametric method, area under the curve(AUC)

Key words: abrupt changes detection,
nonparametric method, area under the curve(AUC)

中图分类号: