一种基于模板匹配与隶属度解析的时序异常快速检测方法

doi:10.16180/j.cnki.issn1007-7820.2022.06.001

摘要/Abstract

摘要：

传统的数据检测技术在处理大规模医疗数据时,耗时较高且抗干扰能力较弱。针对这些问题,文中应用模板匹配与隶属度解析技术,给出了一种时序数据异常状态的快速检测与分析方法。该方法采用TSTKS算法与滑动窗口理论实现时序数据多突变点快速检测,提取连续多窗口波动特征,构建时序数据的归一化波动向量,对大规模病变信号进行异常状态检测与分析。仿真数据与脑电病变信号分析等实验表明,此方法是一种较为快速、准确的大数据分析与检测方法。

关键词: 突变点检测, 大数据分析, 异常检测, 滑动窗口, 时序数据, 波动向量, 模板匹配, 隶属度分析

Abstract:

Traditional data analysis techniques have some problems in processing large-scale medical data detection, such as high time consumption and weak anti-jamming ability. In order to solve these problems, this study presents a fast detection and analysis method for abnormal state of time series data which uses template matching and membership degree analysis techniques. This method uses TSTKS algorithm and sliding window theory to realize rapid detection of multiple mutation points in time series data, extract continuous multi-window fluctuation characteristics, construct a normalized fluctuation vector of time series data, and perform abnormal state detection and analysis on large-scale disease signals. Simulation data and EEG signal analysis show that this method is a relatively fast and accurate method for big data analysis and detection.

Key words: mutation point detection, big data analytics, anomaly detection, sliding window, time series, fluctuating vector, template matching, membership analysis

中图分类号:

TP311

龚汉鑫,齐金鹏,孔凡书,朱俊俊,曹一彤. 一种基于模板匹配与隶属度解析的时序异常快速检测方法[J]. 电子科技, 2022, 35(6): 1-5.

GONG Hanxin,QI Jinpeng,KONG Fanshu,ZHU Junjun,CAO Yitong. A Fast Time Series Anomaly Detection Method Based on Template Matching and Membership Analysis[J]. Electronic Science and Technology, 2022, 35(6): 1-5.

图/表 8

图1

图2

表1

图3

图4

图5

图6

表2

参考文献 19

[1]	Han J S. Utilization outlook of medical big data in the cloud environment[J]. Journal of Digital Convergence, 2014, 12(6):341-347. doi: 10.14400/JDC.2014.12.6.341
[2]	郑秀娟. 数据挖掘技术在医疗大数据中的应用研究[J]. 电脑知识与技术, 2020, 16(32):26-27.
	Zheng Xiujuan. Research on application of data mining technology in medicalbig data[J]. Computer Knowledge and Technology, 2020, 16(32):26-27.
[3]	Miller L H. Table of percentage points of kolmogorov statistics[J]. Journal of the American Statistical Association, 1956, 51(273):111-121. doi: 10.1080/01621459.1956.10501314
[4]	付昱华, 付安捷. 改进的R/S方法与中国火灾数据的分析预测[J]. 中国工程科学, 2004(5):39-44.
	Fu Yuhua, Fu Anjie. Improved R/S method and analysis and forecast to China fire data[J]. Strategic Study of CAE, 2004(5):39-44.
[5]	Qi J P, Zhang Q, Qi J, et al. Multi-channel detection for abrupt change based on the Ternary Search Tree and Kolmogorov statistic method[C]. Hangzhou: Proceedings of the 2015 Thirty-fourth Chinese Control Conference, 2015.
[6]	杨姣, 高仲合, 王来花. 网络数据流异常检测研究[J]. 通信技术, 2019, 52(1):129-133.
	Yang Jiao, Gao Zhonghe, Wang Laihua. Research on network data flow anomaly detection[J]. Communications Technology, 2019, 52(1):129-133.
[7]	马玉超. 基于机器学习的动态基线性能时序数据异常检测研究与应用[J]. 中国金融电脑, 2020(6):51-59.
	Ma Yuchao. Research and application of anomaly detection for performance time-sequence data by dynamic baseline based on machine learning[J]. Financial Computer of China, 2020(6):51-59.
[8]	Song X, Zheng Y, Xue W, et al. Identification of risk genes related to myocardial infarction and the construction of early SVM diagnostic model[J]. International Journal of Cardiology, 2020, 32(8):182-190.
[9]	宋玉龙. 脑电信号中癫痫特征提取方法研究[D]. 镇江: 江苏科技大学, 2019.
	Song Yulong. The study on extraction of epilepsy features in signal[D]. Zhenjiang: Jiangsu University of Science and Technology, 2019.
[10]	朱厚杰. 基于突变点检测的病理信号分析方法研究与应用[D]. 上海: 东华大学, 2020.
	Zhu Houjie. An study and application of pathological signal analysis method based on change point detection[D]. Shanghai: Donghua University, 2020.
[11]	Qi J P, Zhang Q, Zhu Y, et al. A novel method for fast change-point detection on simulated time series and electrocardiogram data[J]. PLoS One, 2014, 9(4):1-15.
[12]	Pearson R K. Outliers in process modeling and identification[J]. IEEE Transactions on Control Systems Technology, 2002, 10(1):55-63. doi: 10.1109/87.974338
[13]	徐玲, 王涛. 一种基于隶属度最大化的组合赋权法[J]. 港航论坛, 2006(2):44-48.
	Xu Ling, Wang Tao. A combined method of determining weight based on maximizing the membership decision function[J]. Harbon and Shipping Forum, 2006(2):44-48.
[14]	冯忠慧. 基于滑动窗口的数据流频繁模式挖掘算法的研究[D]. 天津: 天津工业大学, 2019.
	Feng Zhonghui. Research on frequent pattern mining algorithm of data streambased on sliding window[D]. Tianjin: Tianjin Polytechnic University, 2019.
[15]	邹俊晨, 齐金鹏, 李娜, 等. 一种快速的突变点在线检测算法设计与实现[J]. 电子科技, 2020, 33(8):10-15.
	Zou Junchen, Qi Jinpeng, Li Na, et al. Design and implementation of fast online algorithm for mutation point detection[J]. Electronic Science and Technology, 2020, 33(8):10-15.
[16]	侯丽云, 张旭, 吴珍. 多种归一化方法对miRNA微阵列数据的作用分析及比较[J]. 西南师范大学学报(自然科学版), 2020, 45(5):98-102.
	Hou Liyun, Zhang Xu, Wu Zhen. Analysis and comparison of various normalization methods on microarray data of miRNA[J]. Journal of Southwest China Normal University(Natural Science Edition), 2020, 45(5):98-102.
[17]	吕思奇. 基于整体经验模态分解和随机森林的癫痫脑电信号识别算法研究[D]. 长春: 长春工业大学, 2018.
	Lü Siqi. Feature extraction and recognition of electroencephalogram based on ensemble empirical mode decomposition and random forest[D]. Changchun: Changchun University of Technology, 2018.
[18]	张友迅. 一种先天性心脏病快速筛选仪系统的设计与实现[D]. 南京: 南京邮电大学, 2019.
	Zhang Youxun. Design and implementation of a rapid screening instrument for congenital heart disease[D]. Nanjing: Nanjing University of Posts and Telecommunications, 2019.
[19]	陈建阁. 基于曼哈顿距离法的颗粒流速检测技术[J]. 工矿自动化, 2015, 41(11):52-55.
	Chen Jiange. Particle velocity detection technology based on Manhattan distance method[J]. Industry and Mine Automation, 2015, 41(11):52-55.

窗口宽度W	时耗/ms	误差/%	命中率/%
2⁶	76.6	33.21	62.31
2⁷	34.9	30.69	76.93
2⁸	17.6	26.31	86.56
2⁹	9.1	32.22	87.83
2¹⁰	6.7	31.68	78.57

算法	时耗/ms	准确率/%	查全率/%
隶属度分析法	16.85	88.92	92.33
相关系数法	23.47	85.69	87.58
Manhattan距离法	14.23	76.82	68.35