Symbolic representation based on trend features for knowledge discovery in long time series*

Yin Hong; Yang Shu-qiang; Zhu Xiao-qian; Ma Shao-dong; Zhang Lu-min

doi:10.1631/FITEE.1400376

Your Location：

Home >

Browse articles >

Symbolic representation based on trend features for knowledge discovery in long time series*

Computer & Automation | Updated：2022-06-06

- Symbolic representation based on trend features for knowledge discovery in long time series*
- 基于趋势特征的时间序列符号化方法
- Frontiers of Information Technology & Electronic Engineering Vol. 16, Issue 9, Pages: 744-758(2015)
- Affiliations：
  
  1.College of Computer, National University of Defense Technology, Changsha 410073, China
  2.School of Engineering, University of Hull, Cottingham Road HU6 7RX, UK
  3.Xiangyang School for NCOs, Xiangyang 441118, China
- Author bio：
  
  †E-mail:yinhonggfkd@aliyun.com
- Funds：
- DOI：10.1631/FITEE.1400376
  CLC：
- Received：02 November 2014，
  
  Revised：2015-08-10，
  
  Accepted：27 May 2015，
  
  Published：2015-09
- Accepted：
Scan QR Code
Yin Hong, Yang Shu-qiang, Zhu Xiao-qian, et al. Symbolic representation based on trend features for knowledge discovery in long time series*[J]. Frontiers of Information Technology & Electronic Engineering, 2015, 16(9): 744-758.
DOI：

Yin Hong, Yang Shu-qiang, Zhu Xiao-qian, et al. Symbolic representation based on trend features for knowledge discovery in long time series*[J]. Frontiers of Information Technology & Electronic Engineering, 2015, 16(9): 744-758. DOI： 10.1631/FITEE.1400376.

摘要

目的

提出一种通用方法用于长时间序列的知识发现过程。

创新点

提出一种基于并行分割的时间序列符号化方法—趋势特征符号化近似法（trend feature symbolic approximation

TFSA），对长时间序列进行快速分割，并且保留原始序列大多数趋势特征，将分割后的子序列用特征符号表示。本文的贡献在于改进了长时间序列的分割效率，而且TFSA专注于保留原始时间序列的大多数趋势特征，使得挖掘后的规则更加容易理解和解释。

方法

首先，通过一个两步（two-step）分割机制将时间序列分割成一系列不等长的子序列。然后，采用趋势特征符号化近似（TFSA）将子序列符号化并获得符号项集。最后通过一个基于apriori的关联规则算法来实现时序数据的知识发现。

结论

针对长时间序列，基于累积和控制图方法研究一种海量数据环境下序列的并行分割机制。可以通过分布式结点来实现，随结点数增加，其效率将进一步提高。TFSA符号化方法不同于传统的方法，它致力于保留原始时间序列的大部分趋势特征及模式，通过规定的趋势符号来表示时间序列，并且其表达方式也考虑后续的时间序列挖掘研究。实验证明，本文方法在时间序列的分割效率以及分类准确性上相比于已有的方法均有所提高。

Abstract

The symbolic representation of time series has attracted much research interest recently. The high dimensionality typical of the data is challenging

especially as the time series becomes longer. The wide distribution of sensors collecting more and more data exacerbates the problem. Representing a time series effectively is an essential task for decision-making activities such as classification

prediction

and knowledge discovery. In this paper

we propose a new symbolic representation method for long time series based on trend features

called trend feature symbolic approximation (TFSA). The method uses a two-step mechanism to segment long time series rapidly. Unlike some previous symbolic methods

it focuses on retaining most of the trend features and patterns of the original series. A time series is represented by trend symbols

which are also suitable for use in knowledge discovery

such as association rules mining. TFSA provides the lower bounding guarantee. Experimental results show that

compared with some previous methods

it not only has better segmentation efficiency and classification accuracy

but also is applicable for use in knowledge discovery from time series.

关键词

Keywords

references

R Agrawal , R Srikant . Mining sequential patterns . 1995 . Proc. 11th Int. Conf. on Data Engineering . 3 - 14 . DOI: 10.1109/ICDE.1995.380415 http://doi.org/10.1109/ICDE.1995.380415 .

H André-Jönsson , DZ Badal . Using signature files for querying time-series data . 1997 . Proc. 1st European Symp. on Principles of Data Mining and Knowledge Discovery . 211 - 220 . DOI: 10.1007/3-540-63223-9_120 http://doi.org/10.1007/3-540-63223-9_120 .

D Bao , Z Yang . Intelligent stock trading system by turning point confirming and probabilistic reasoning . Expert Syst Appl , 2008 . 34 ( 1 ): 620 - 627 . DOI: 10.1016/j.eswa.2006.09.043 http://doi.org/10.1016/j.eswa.2006.09.043 .

C Borgelt , R Kruse . Induction of association rules: apriori implementation . 2002 . Proc. Computational Statistics . 395 - 400 . DOI: 10.1007/978-3-642-57489-4_59 http://doi.org/10.1007/978-3-642-57489-4_59 .

Y Bu , L Chen , AWC Fu , 等 . Efficient anomaly monitoring over moving object trajectory streams . 2009 . Proc. 15th ACM SIGKDD Int. Conf. on Knowledge Discovery and Data Mining . 159 - 168 . DOI: 10.1145/1557019.1557043 http://doi.org/10.1145/1557019.1557043 .

KP Chan , AWC Fu . Efficient time series matching by wavelets . 1999 . Proc. 15th Int. Conf. on Data Engineering . 126 - 133 . DOI: 10.1109/ICDE.1999.754915 http://doi.org/10.1109/ICDE.1999.754915 .

D Dasgupta , S Forrest . Novelty detection in time series data using ideas from immunology . 1996 . Proc. 5th Int. Conf. on Intelligent Systems . 82 - 87 . .

P Esling , C Agon . Time-series data mining . ACM Comput Surv, 45 (1), Article 12 , 2012 . DOI: 10.1145/2379776.2379788 http://doi.org/10.1145/2379776.2379788 .

C Faloutsos , M Ranganathan , Y Manolopoulos . Fast subsequence matching in time-series databases . 1994 . Proc. ACM SIGMOD Int. Conf. on Management of Data . 419 - 429 . DOI: 10.1145/191839.191925 http://doi.org/10.1145/191839.191925 .

G Guimarães , A Ultsch . A method for temporal knowledge conversion . 1999 . Proc. 3rd Int. Symp. on Advances in Intelligent Data Analysis . 369 - 380 . DOI: 10.1007/3-540-48412-4_31 http://doi.org/10.1007/3-540-48412-4_31 .

G Guimarães , JH Peter , T Penzel , 等 . A method for automated temporal knowledge acquisition applied to sleep-related breathing disorders . Artif Intell Med , 2001 . 23 ( 3 ): 211 - 237 . DOI: 10.1016/S0933-3657(01)00089-6 http://doi.org/10.1016/S0933-3657(01)00089-6 .

MW Kadous . Learning comprehensible descriptions of multivariate time series . 1999 . Proc. 16th Int. Conf. of Machine Learning . 454 - 463 . .

E Keogh , K Chakrabarti , M Pazzani , 等 . Locally adaptive dimensionality reduction for indexing large time series databases . 2001 . Proc. ACM SIGMOD Int. Conf. on Management of Data . 151 - 162 . DOI: 10.1145/375663.375680 http://doi.org/10.1145/375663.375680 .

M Kontaki , AN Papadopoulos , Y Manolopoulos . Continuous trend-based classification of streaming time series . 2005 . Proc. 9th East European Conf. on Advances in Databases and Information Systems . 294 - 308 . DOI: 10.1007/11547686_22 http://doi.org/10.1007/11547686_22 .

M Kontaki , AN Papadopoulos , Y Manolopoulos . Continuous trend-based clustering in data streams . 2008 . Proc. 10th Int. Conf. on Data Warehousing and Knowledge Discovery . 251 - 262 . DOI: 10.1007/978-3-540-85836-2_24 http://doi.org/10.1007/978-3-540-85836-2_24 .

F Korn , HV Jagadish , C Faloutsos . Efficiently supporting ad hoc queries in large datasets of time sequences . 1997 . Proc. ACM SIGMOD Int. Conf. on Management of Data . 289 - 300 . DOI: 10.1145/253260.253332 http://doi.org/10.1145/253260.253332 .

M Lavielle , G Teyssière . Detection of multiple change-points in multivariate time series . Lithuan Math J , 2006 . 46 ( 3 ): 287 - 306 . DOI: 10.1007/s10986-006-0028-9 http://doi.org/10.1007/s10986-006-0028-9 .

J Lin , E Keogh , S Lonardi , 等 . A symbolic representation of time series, with implications for streaming algorithms . 2003 . Proc. 8th ACM SIGMOD Workshop on Research Issues in Data Mining and Knowledge Discovery . 2 - 11 . DOI: 10.1145/882082.882086 http://doi.org/10.1145/882082.882086 .

S Manganaris . Supervised Classification with Temporal Data. PhD T , 1997 . Vanderbilt University, USA , .

H Mannila , H Toivonen . Discovering generalized episodes using minimal occurrences . 1996 . Proc. Int. Conf. on Knowledge Discovery and Data Mining . 146 - 151 . .

A Mellit , AM Pavan , M Benghanem . Least squares support vector machine for short-term prediction of meteorological time series . Theor Appl Climatol , 2013 . 111 ( 1-2 ): 297 - 307 . DOI: 10.1007/s00704-012-0661-7 http://doi.org/10.1007/s00704-012-0661-7 .

GB Moody , RG Mark . A new method for detecting atrial fibrillation using RR intervals . Comput Cardiol , 1983 . 10 227 - 230 . .

C Phetking , M Noor Md Sap , A Selamat . A multiresolution important point retrieval method for financial time series representation . 2008 . Proc. Int. Conf. on Computer and Communication Engineering . 510 - 515 . DOI: 10.1109/ICCCE.2008.4580656 http://doi.org/10.1109/ICCCE.2008.4580656 .

S Poll , J de Kleer , A Feldman , 等 . Second international diagnostics competition—DXC’10 . 2010 . Proc. 21st Int. Workshop on Principles of Diagnosis . 1 - 15 . .

S Sarkar , K Mukherjee , S Sarkar , 等 . Symbolic dynamic analysis of transient time series for fault detection in gas turbine engines . J Dynam Syst, Meas Contr , 2013 . 135 ( 1 ): 014506.1 - 0145066 . DOI: 10.1115/1.4007699 http://doi.org/10.1115/1.4007699 .

R Villafane , KA Hua , D Tran , 等 . Knowledge discovery from series of interval events . J Intell Inform Syst , 2000 . 15 ( 1 ): 71 - 89 . DOI: 10.1023/A:1008781812242 http://doi.org/10.1023/A:1008781812242 .

HJLM Vullings , MHG Verhaegen , HB Verbruggen . ECG segmentation using time-warping . 1997 . Proc. 2nd Int. Symp. on Advances in Intelligent Data Analysis Reasoning about Data . 275 - 285 . DOI: 10.1007/BFb0052847 http://doi.org/10.1007/BFb0052847 .

AB Yeh , DKJ Lin , C Venkataramani . Unified CUSUM charts for monitoring process mean and variability . Qual Technol Quant Manag , 2004 . 1 ( 1 ): 65 - 86 . .

Views

Downloads

CSCD

Alert me when the article has been cited

Submit

Tools

Publicity Resources

No data

Related Author

No data

Related Institution

No data

Chat

Address：Zhejiang University Press, 148 Tianmushan Road, Hangzhou, China Postal code：310028
Tel：+86-571-88273162 Email：fitee@zju.edu.cn
It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.

⁰