Time Series Epenthesis: Clustering Time Series Streams Requires Ignoring Some Data

Rakthanmanon, Thanawin; Keogh, Eamonn; Lonardi, Stefano; Evans, S.

doi:10.1109/icdm.2011.146

Cited by 96 publications

(96 citation statements)

References 24 publications

(43 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The work closest to ours in spirit is that of Rakthanmanon et al, which shows the importance of ignoring some data for clustering within a single time series stream [23].…”

Section: Related Workmentioning

confidence: 85%

Clustering Time Series Using Unsupervised-Shapelets

Zakaria

Mueen

Keogh

2012

2012 IEEE 12th International Conference on Data Mining

165

103

View full text Add to dashboard Cite

Abstract-Time series clustering has become an increasingly important research topic over the past decade. Most existing methods for time series clustering rely on distances calculated from the entire raw data using the Euclidean distance or Dynamic Time Warping distance as the distance measure. However, the presence of significant noise, dropouts, or extraneous data can greatly limit the accuracy of clustering in this domain. Moreover, for most real world problems, we cannot expect objects from the same class to be equal in length. As a consequence, most work on time series clustering only considers the clustering of individual time series "behaviors," e.g., individual heart beats or individual gait cycles, and contrives the time series in some way to make them all equal in length. However, contriving the data in such a way is often a harder problem than the clustering itself.In this work, we show that by using only some local patterns and deliberately ignoring the rest of the data, we can mitigate the above problems and cluster time series of different lengths, i.e., cluster one heartbeat with multiple heartbeats. To achieve this we exploit and extend a recently introduced concept in time series data mining called shapelets. Unlike existing work, our work demonstrates for the first time the unintuitive fact that shapelets can be learned from unlabeled time series. We show, with extensive empirical evaluation in diverse domains, that our method is more accurate than existing methods. Moreover, in addition to accurate clustering results, we show that our work also has the potential to give insights into the domains to which it is applied.

show abstract

“…The work closest to ours in spirit is that of Rakthanmanon et al, which shows the importance of ignoring some data for clustering within a single time series stream [23].…”

Section: Related Workmentioning

confidence: 85%

Clustering Time Series Using Unsupervised-Shapelets

Zakaria

Mueen

Keogh

2012

2012 IEEE 12th International Conference on Data Mining

165

103

View full text Add to dashboard Cite

show abstract

“…However, the size of the subsequences has to be predefined and only the subsequences with similar length can be compared, as they use the Euclidean distance. Rakthanmanon et al [15] introduced the idea, which we also applied, that not all the data should be clustered and transition phases should be ignored. They developed a clustering algorithm for discrete time series, based on the Minimum Description Length principle.…”

Section: State Anomaly Detectionmentioning

confidence: 99%

Online unsupervised state recognition in sensor data

Eberle

Wijaya

Aberer

2015

2015 IEEE International Conference on Pervasive Computing and Communications (PerCom)

View full text Add to dashboard Cite

Abstract-Smart sensors, such as smart meters or smart phones, are nowadays ubiquitous. To be "smart", however, they need to process their input data with limited storage and computational resources. In this paper, we convert the stream of sensor data into a stream of symbols, and further, to higher level symbols in such a way that common analytical tasks such as anomaly detection, forecasting or state recognition, can still be carried out on the transformed data with almost no loss of accuracy, and using far fewer resources. We identify states of a monitored system and convert them into symbols (thus, reducing data size), while keeping "interesting" events, such as anomalies or transition between states, as it is. Our algorithm is able to find states of various length in an online and unsupervised way, which is crucial since behavior of the system is not known beforehand. We show the effectiveness of our approach using real-world datasets and various application scenarios.

show abstract

“…The template in the left picture of Fig. 1 stands for a piece of bird song [45], which helps us find recurring subsequences from the time series in the right picture of Fig. 1. …”

Section: Definitionmentioning

confidence: 99%

“…It has been pointed out by researchers that some unspecified portions of time series should be ignored [3,45] to achieve a better matching result, which means some data points have nothing to do with predefined patterns, and should be filtered out. Ye and Keogh [57] propose a new time series primitive, time series shapelets, for time series classification.…”

Section: Related Workmentioning

confidence: 99%

Predefined pattern detection in large time series

Miao

Vespier

Cachucho

et al. 2016

Information Sciences

View full text Add to dashboard Cite

a b s t r a c tPredefined pattern detection from time series is an interesting and challenging task. In order to reduce its computational cost and increase effectiveness, a number of time series representation methods and similarity measures have been proposed. Most of the existing methods focus on full sequence matching, that is, sequences with clearly defined beginnings and endings, where all data points contribute to the match. These methods, however, do not account for temporal and magnitude deformations in the data and result to be ineffective on several real-world scenarios where noise and external phenomena introduce diversity in the class of patterns to be matched. In this paper, we present a novel pattern detection method, which is based on the notions of templates, landmarks, constraints and trust regions. We employ the Minimum Description Length (MDL) principle for time series preprocessing step, which helps to preserve all the prominent features and prevents the template from overfitting. Templates are provided by common users or domain experts, and represent interesting patterns we want to detect from time series. Instead of utilising templates to match all the potential subsequences in the time series, we translate the time series and templates into landmark sequences, and detect patterns from landmark sequence of the time series. Through defining constraints within the template landmark sequence, we effectively extract all the landmark subsequences from the time series landmark sequence, and obtain a number of landmark segments (time series subsequences or instances). We model each landmark segment through scaling the template in both temporal and magnitude dimensions. To suppress the influence of noise, we introduce the concept of trust region, which not only helps to achieve an improved instance model, but also helps to catch the accurate boundaries of instances of the given template. Based on the similarities derived from instance models, we introduce the probability density function to calculate a similarity threshold. The threshold can be used to judge if a landmark segment is a true instance of the given template or not. To evaluate the effectiveness and efficiency of the proposed method, we apply it to two real-world datasets. The results show that our method is capable of detecting patterns of temporal and magnitude deformations with competitive performance.

show abstract

Time Series Epenthesis: Clustering Time Series Streams Requires Ignoring Some Data

Cited by 96 publications

References 24 publications

Clustering Time Series Using Unsupervised-Shapelets

Clustering Time Series Using Unsupervised-Shapelets

Online unsupervised state recognition in sensor data

Predefined pattern detection in large time series

Contact Info

Product

Resources

About