An efficient and accurate method for evaluating time series similarity

Morse, Michael D.; Patel, Jignesh M.

doi:10.1145/1247480.1247544

Cited by 146 publications

(93 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To address this question, we also investigated methods that could enhance the computational speed of the LCSS algorithm. We thus investigate the Fast Time Series Evaluation algorithm (FTSE) [27] in an attempt to make LCSS faster. LCSS in conjunction with FTSE claims to be faster than LCSS algorithm alone because unlike LCSS calculations, FTSE does not use a two dimensional array to compare matches between two time series.…”

Section: Attempt To Improve Computational Speed: Fast Time Series mentioning

confidence: 99%

“…In this case, the element B of Y is found to match the second grid cell and thus the value inside the cell enters the intersection list. Likewise, the other elements of Y are also probed in a similar manner and the total matching elements between the sequences are recorded in the intersection list [27]. In this example, the contributing indices 2, 3, 4, 6 gives the subsequence, viz., B, C, B, A.…”

Section: Attempt To Improve Computational Speed: Fast Time Series mentioning

confidence: 99%

“…The combined FTSE +LCSS became rapidly more expensive with increasing sample size. The computation of LCSS from the intersection list of FTSE shown in figure 2 is carried out in three nested loops [27]. The first loop is used to go to individual cells of the intersection list.…”

Section: Conclusion and Future Directionmentioning

confidence: 99%

“…The space complexity of LCSS+FTSE is very high compared to LCSS alone and most of the grid cells of the former are usually unoccupied. The amount of space could be reduced by increasing the threshold value [27] (reducing the fineness of the grid) which, unfortunately would compromise the accuracy of LCSS. In the future applications of the combined FTSE and LCSS, the algorithms needs to be efficiently parallelized for treating large sample sizes.…”

Section: Conclusion and Future Directionmentioning

confidence: 99%

“…The current paper explores fast, accurate and efficient methods for unsupervised classification of trigger waveforms further. An analysis pipeline has been constructed based on LCSS and also LCSS in conjunction with Fast Time Series Evaluation (FTSE) [27]. The latter is done to explore possibilities of increasing computational speed when very large datasets of triggers are involved.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

New approach to time domain classification of broadband noise in gravitational wave data

Mukherjee¹,

Rizwan²,

Biswas³

2012

Phys. Rev. D

View full text Add to dashboard Cite

Transient broadband noise in gravitational wave (GW) detectors, also known as noise triggers (referred to as triggers for brevity), can often be a deterrant to the efficiency with which astrophysical search pipelines detect sources. It is important to understand their instrumental or environmental origin so that they could be eliminated or accounted for in the data. Since the number of triggers is large, data mining approaches such as clustering and classification are useful tools for this task. Classification of triggers based on a handful of discrete properties has been done in the past. A rich information content is available in the waveform or 'shape' of the triggers that has had a rather restricted exploration so far. This paper presents a new way to classify triggers deriving information from both trigger waveforms as well as their discrete physical properties using a sequential combination of the Longest Common Sub-Sequence (LCSS) and LCSS coupled with Fast Time Series Evaluation (FTSE) for waveform classification and the multidimensional hierarchical classification (MHC) analysis for the grouping based on physical properties. A generalized k-means algorithm is used with the LCSS (and LCSS+FTSE) for clustering the triggers using a validity measure to determine the correct number of clusters in absence of any prior knowledge. The results have been demonstrated by simulations and by application to a segment of real LIGO data from the sixth science run.

show abstract

Section: Attempt To Improve Computational Speed: Fast Time Series mentioning

confidence: 99%

Section: Attempt To Improve Computational Speed: Fast Time Series mentioning

confidence: 99%

Section: Conclusion and Future Directionmentioning

confidence: 99%

Section: Conclusion and Future Directionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

New approach to time domain classification of broadband noise in gravitational wave data

Mukherjee¹,

Rizwan²,

Biswas³

2012

Phys. Rev. D

View full text Add to dashboard Cite

show abstract

A marginal contribution coefficient for sequences of nonstationary continuous Markov chains

Souza

Duarte

Mendes

2017

Appl Stoch Models Bus & Ind

View full text Add to dashboard Cite

In this work, a set of sequences of information (time series), under nonstationary regime, with continuous space state, discrete time, and a Markovian dependence, is considered. A new model that expresses the marginal transition density function of one sequence as a linear combination of the marginal transition density functions of all sequences in the set is proposed. The coefficients of this combination are denominated marginal contribution coefficients and represent how much each transition density function contributes to the calculation of a chosen transition density function. The proposed coefficient is a marginal coefficient because it can be computed instantaneously, and it may change from one time to another time since all calculations are performed before stationarity is reached. This clearly differentiates the new coefficient from well‐known measures such as the cross‐correlation and the coherence. The idea behind the model is that if a specific sequence has a high marginal contribution for the transition density function from another sequence, the first may be replaced by the latter without losing much information that means that the knowledge of few densities should be enough to recover the overall behaviour. Simulations, considering 2 chains, are presented so as to check the sensitivity of the proposed model. The methodology is also applied to a real data originated from a wire‐drawing machine whose main function is to decrease the transverse diameter of metal wires. The behaviour of the level of acceleration of each bearing in relation to the other ones is then verified.

show abstract

Performance evaluation of reanalysis precipitation products in Egypt using fuzzy entropy time series similarity analysis

Hamed

Nashwan

Shahid

2021

Intl Journal of Climatology

View full text Add to dashboard Cite

Selection of suitable gridded precipitation data is deemed for hydroclimatic assessment and climate change impact analysis, especially in regions where long‐term reliable precipitation data is unavailable. A novel approach based on fuzzy entropy similarity analysis (FESA) is proposed to evaluate the performance of four reanalysis gridded precipitation datasets (GPDs) for Egypt, namely European Reanalysis v.5. (ERA5), TerraClimate, Global Land Data Assimilation System (GLDAS)—Noah Land Surface Model L4 v.2 and Climatologies at high resolution for the Earth's land surface areas (CHELSA), against gauge records. The proposed method was verified using conventional statistics. Besides, the relative performance of different GPDs was verified according to their response to the influence of North Atlantic Oscillation (NAO) and Mediterranean Oscillation (MO) on winter precipitation. The performance of the best reanalysis GPD was also compared with the gauge‐based global precipitation climatology centre (GPCC) dataset to show its reliability. The FESA revealed CHELSA as the best reanalysis GPD for Egypt. The performance assessment of GDPs based on conventional statistical metrics and visual presentation confirms the results obtained using FESA. CHELSA showed significant correlations with NAO (r = 0.3627) and MO (r = 0.624) like that obtained for gauge records. CHELSA also showed a better representation of precipitation in Egypt than GPCC at nearly half of the gauge locations. As CHELSA has a much higher spatial resolution than GPCC, it can be recommended as the proxy of gauge records in Egypt. The FESA can be used for performance analysis of gridded climate data by avoiding the complexities of using multiple statistical metrics.

show abstract

An efficient and accurate method for evaluating time series similarity

Cited by 146 publications

References 21 publications

New approach to time domain classification of broadband noise in gravitational wave data

New approach to time domain classification of broadband noise in gravitational wave data

A marginal contribution coefficient for sequences of nonstationary continuous Markov chains

Performance evaluation of reanalysis precipitation products in Egypt using fuzzy entropy time series similarity analysis

Contact Info

Product

Resources

About