Online amnesic approximation of streaming time series

Palpanas, Themis; Vlachos, Michail; Keogh, Eamonn; Gunopulos, Dimitrios; Truppel, Wagner

doi:10.1109/icde.2004.1320009

Cited by 99 publications

(92 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Here, our primary objective was to demonstrate how model-based techniques are used for improving various aspects of query processing over sensor data. Lastly, we discussed data compression techniques, like, linear approximation [34,39,48], multi-model approximations [39,40,50] and orthogonal transformations [1,22,53,7].…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

A Survey of Model-based Sensor Data Acquisition and Management

Sathe

Papaioannou

Jeung³

et al. 2012

Managing and Mining Sensor Data

View full text Add to dashboard Cite

In recent years, due to the proliferation of sensor networks, there has been a genuine need of researching techniques for sensor data acquisition and management. To this end, a large number of techniques have emerged that advocate model-based sensor data acquisition and management. These techniques use mathematical models for performing various, day-to-day tasks involved in managing sensor data. In this chapter, we survey the state-of-the-art techniques for model-based sensor data acquisition and management. We start by discussing the techniques for acquiring sensor data. We, then, discuss the application of models in sensor data cleaning; followed by a discussion on model-based meth-

show abstract

Section: Discussionmentioning

confidence: 99%

“…Palpanas et al [48] employ amnesic functions and propose novel techniques that are applicable to a wide range of user-defined approximating functions. According to amnesic functions, recent data is approximated with higher accuracy, while higher error can be tolerated for older data.…”

Section: Swing and Slide Filtersmentioning

confidence: 99%

A Survey of Model-based Sensor Data Acquisition and Management

Sathe

Papaioannou

Jeung³

et al. 2012

Managing and Mining Sensor Data

View full text Add to dashboard Cite

show abstract

“…In this example we compare our approach to the most referenced method [81], which uses PLA. We found that even if we force the motif-representation based method to use half the space of PLA, it can still approximate the data with a residual error that is approximately one-ninth that of PLA.…”

Section: Online Summarization/compressionmentioning

confidence: 99%

Exact Primitives for Time Series Data Mining

Mueen

2012

Proceedings of the 18th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

View full text Add to dashboard Cite

“…The shapelets are informally defined as the subsequences that are in some sense maximally representative of a class. This method is interpretable and accurate in classifying static time series [42], but is ineffective in handling real time time series. Inspired by these works, we introduce trust region into time series to obtain more reliable instance models.…”

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

Online amnesic approximation of streaming time series

Abstract: Abstract

Cited by 99 publications

References 24 publications

A Survey of Model-based Sensor Data Acquisition and Management

A Survey of Model-based Sensor Data Acquisition and Management

Exact Primitives for Time Series Data Mining

Predefined pattern detection in large time series

Contact Info

Product

Resources

About