Time Series Classification Based on Spectral Analysis

Jeng, Shuen-Lin; Huang, Ya Ti

doi:10.1080/03610910701723971

Cited by 11 publications

(8 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The same representation was used in [Geurts 2001]; it is apparently not too robust to outliers. To overcome the obstacle of high dimensionality, [Jeng and Huang 2008] used Singular Value Decomposition to select essential frequencies. However, it implies higher computational costs.…”

Section: Classificationmentioning

confidence: 99%

Time-series data mining

2012

View full text Add to dashboard Cite

In almost every scientific field, measurements are performed over time. These observations lead to a collection of organized data called time series. The purpose of time series data mining is to try to extract all meaningful knowledge from the shape of data. Even if humans have a natural capacity to perform these tasks, it remains a complex problem for computers. In this paper we intend to provide a survey of the techniques applied for time series data mining. The first part is devoted to an overview of the tasks that have captured most of the interest of researchers. Considering that in most cases, time series task relies on the same components for implementation, we divide the literature depending on these common aspects, namely representation techniques, distance measures and indexing methods. The study of the relevant literature has been categorized for each individual aspects. Four types of robustness could then be formalized and any kind of distance could then be classified. Finally, the study submit various research trends and avenues that can be explored in the near future. We hope that this paper can provide a broad and deep understanding of the time series data mining research field.

show abstract

Section: Classificationmentioning

confidence: 99%

Time-series data mining

2012

View full text Add to dashboard Cite

show abstract

“…Spectrum analysis has become a popular frequency-domain method for a sampled process, which can dig out corresponding relations of frequency components and amplitude among each signal and can find the most important part hidden in the signal series [1,2] . Based on the concept of information theory, Burg [3] proposed maximum entropy spectral analysis (MESA) to estimate frequency spectral.…”

Section: Ntroductionmentioning

confidence: 99%

Maximum Entropy Spectral estimation based on accelerating genetic algorithm

Ming

Zhang

Jin

et al. 2009

2009 Chinese Control and Decision Conference

View full text Add to dashboard Cite

The purpose of this paper was to solve the problems of spectral peak shifting and line splitting existing in Burg's Maximum Entropy Spectral Analysis method (MESA), to enhance the resolution of entropy spectral, and to increase the adaptability of spectral estimation algorithm to signal length, signal noise ratio and initial phase. A method of accelerating Genetic algorithm based maximum Entropy Spectral estimation method (GES) was proposed, where accelerating genetic algorithm was used to optimize the parameters of MESA and the four equivalent conditions of MESA were used as objective function. Three typical simulation cases indicated that the phenomenon of spectral peak shifting and line splitting were absent in the frequency spectral estimated by GES, and the ability to discriminate two closed frequency was improved. Compared with the traditional MESA methods, GES has good performances in signal processing.

show abstract

“…Figure 1 shows a typical earthquake and a mining explosion, as well as an event at Novaya Zemlya, Russia, from unknown origin (Kakizawa et al ., 1998; Shumway and Stoffer, 2006; Jeng and Huang, 2008; Fryzlewicz and Ombao, 2009). The known events occurred in the Scandinavian peninsula and were recorded by seismic arrays located in Norway by Norwegian and Arctic experimental seismic stations (NORESS, ARCESS) and in Finland by Finnish experimental seismic stations (FINESS).…”

Section: An Application To Earthquakes and Explosions Datamentioning

confidence: 99%

“…The problem of identifying similarities or dissimilarities in time‐series data has been studied by several authors (Piccolo, 1990; Galeano and Peña, 2000; Maharaj, 2002; Caiado et al ., 2006; Jeng and Huang, 2008). One fundamental issue is the choice of a relevant metric in which the discriminant analysis will be based.…”

Section: Introductionmentioning

confidence: 99%

Time‐series clustering via quasi U‐statistics

Valk

Pinheiro

2012

Journal Time Series Analysis

View full text Add to dashboard Cite

The problem of time-series discrimination and classification is discussed. We propose a novel clustering algorithm based on a class of quasi U-statistics and subgroup decomposition tests. The decomposition may be applied to any concave time-series distance. The resulting test statistics are proven to be asymptotically normal for either i.i.d. or non-identically distributed groups of time-series under mild conditions. We illustrate its empirical performance on a simulation study and a real data analysis. The simulation setup includes stationary vs. stationary and stationary vs. non-stationary cases. The performance of the proposed method is favourably compared with some of the most common clustering measures available.

show abstract

Time Series Classification Based on Spectral Analysis

Cited by 11 publications

References 10 publications

Time-series data mining

Time-series data mining

Maximum Entropy Spectral estimation based on accelerating genetic algorithm

Time‐series clustering via quasi U‐statistics

Contact Info

Product

Resources

About