Maximum error-bounded Piecewise Linear Representation for online stream approximation

Xie, Qing; Pang, Chaoyi; Zhou, Xiaofang; Zhang, Xiangliang; Deng, Ke

doi:10.1007/s00778-014-0355-0

Cited by 75 publications

(52 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since Alg needs more computation by updating the extremes lines, the cost is the most one. As paper [3], we do some experiment on open data sets. The data sets contain a large number of real data with diverse characteristics, and it widely used to evaluate time series problems.…”

Section: Comparisonmentioning

confidence: 99%

See 1 more Smart Citation

An extensive experimental study on segmenting online time series with error bound guarantee

Liu

Wang

Guo

et al. 2015

Int. J. Mach. Learn. & Cyber.

View full text Add to dashboard Cite

show abstract

Section: Comparisonmentioning

confidence: 99%

“…Qi et al [2] extend FS to the polynomial functions and process multidimensional data. For the disconnected one, Xie et al [3] give an optimal algorithm (denoted by DisAlg in this paper), which is designed by maximizing the local segments. Furthermore, another strategy (denoted by Alg) to segment connected lines is discussed in the paper.…”

Section: Introductionmentioning

confidence: 99%

An extensive experimental study on segmenting online time series with error bound guarantee

Liu

Wang

Guo

et al. 2015

Int. J. Mach. Learn. & Cyber.

View full text Add to dashboard Cite

show abstract

“…In practical applications, the dynamics of data such as trends, patterns and outliers are most attractive. In this regard, many techniques such as histogram, line segment and wavelet are employed for effective data approximation and efficient stream query processing [9,11,12,32].…”

Section: Introductionmentioning

confidence: 99%

“…The use of line segments to represent a time series data stream, termed as Piecewise Linear Representation (PLR) [4,6,16,32], has been extensively studied for decades under different criteria [22]. The idea of PLR is to represent a complicated wave-like data stream with a number of simple line segments, so that the streaming data can be efficiently archived, and a query on the stream can be approximately answered by a query on the line segments.…”

Section: Introductionmentioning

confidence: 99%

“…In this chapter, we introduce the state-of-art optimal algorithms to generate the error-bounded PLR for data stream, which aim to construct the smallest number of line segments with approximation error constrained on each data point. We will investigate the algorithm designed in time domain named as OptimalPLR [32], and also another algorithm proposed in an interesting parameter space, which is named as ParaOptimal [20]. Such innovative transformed space has been applied in the recent research for advanced applications [27,34].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Error-Bounded Approximation of Data Stream: Methods and Theories

Xie

Pang

Zhou

et al. 2018

Studies in Big Data

Self Cite

View full text Add to dashboard Cite

As the development of sensor network and Internet of Things, the volume of data is rapidly increasing and the streaming data has attracted much attention recently. To efficiently process and explore data streams, the compact data representation is playing an important role, since the data approximations other than the original data items are usually applied in many stream mining tasks, such as clustering, classification and correlation analysis. In this chapter, we focus on the maximum error-bounded approximation of data stream, which represents the streaming data with constrained approximation error on each data point. There are two criteria for the approximation solution: self-adaption over time for varied error bound and real-time processing. We reviewed the existing data approximation techniques and summarized some essential theories such as optimization guarantee. Two op-Qing Xie( ) * Reprinted by permission from Springer Nature: Springer, The VLDB Journal, Maximum error-bounded piecewise linear representation for online stream approximation, Q. Xie et al., c Springer-Verlag Berlin Heidelberg 2014. (doi: https://doi.org/10.1007/s00778-014-0355-0) 2 Q. Xie et al.timal linear-time algorithms are introduced to construct error-bounded piecewise linear representation for data stream. One generates the line segments by data convex analysis, and the other one is based on the transformed space, which can be extended to a general model. We theoretically analyzed and compared these two different spaces, and proved the theoretical equivalence between them, as well as the two algorithms.

show abstract

Complex brain activity analysis and recognition based on multiagent methods

Zhang

Liu

Dowens

2020

Concurrency and Computation

View full text Add to dashboard Cite

Summary Brain activity recognition research has been a challenging area for many decades since Hans Berger described electroencephalogram (EEG) in 1929. Many previous researches cannot successfully identify EEG status due to dynamic brain activities and complicated brain correlation. This article adopts multiagent‐based methods to analyze EEG datasets, which can enhance the analytical efficiency through incorporating autonomous, self‐coordination characteristics of agents. Intelligent agents are autonomous applications that can improve system compatibility. The preliminary results indicate that the combination of time‐dependency correlation method with multiagent method is an efficient solution for brain activity recognition.

show abstract

Maximum error-bounded Piecewise Linear Representation for online stream approximation

Cited by 75 publications

References 25 publications

An extensive experimental study on segmenting online time series with error bound guarantee

An extensive experimental study on segmenting online time series with error bound guarantee

Error-Bounded Approximation of Data Stream: Methods and Theories

Complex brain activity analysis and recognition based on multiagent methods

Contact Info

Product

Resources

About