Powered by OpenAIRE graph
Found an issue? Give us feedback
image/svg+xml Jakob Voss, based on art designer at PLoS, modified by Wikipedia users Nina and Beao Closed Access logo, derived from PLoS Open Access logo. This version with transparent background. http://commons.wikimedia.org/wiki/File:Closed_Access_logo_transparent.svg Jakob Voss, based on art designer at PLoS, modified by Wikipedia users Nina and Beao Recolector de Cienci...arrow_drop_down
image/svg+xml Jakob Voss, based on art designer at PLoS, modified by Wikipedia users Nina and Beao Closed Access logo, derived from PLoS Open Access logo. This version with transparent background. http://commons.wikimedia.org/wiki/File:Closed_Access_logo_transparent.svg Jakob Voss, based on art designer at PLoS, modified by Wikipedia users Nina and Beao
image/svg+xml Jakob Voss, based on art designer at PLoS, modified by Wikipedia users Nina and Beao Closed Access logo, derived from PLoS Open Access logo. This version with transparent background. http://commons.wikimedia.org/wiki/File:Closed_Access_logo_transparent.svg Jakob Voss, based on art designer at PLoS, modified by Wikipedia users Nina and Beao
versions View all 2 versions
addClaim

States changes representation for time series

Authors: García Pavioni, Alihuén;

States changes representation for time series

Abstract

CAT- El creixement exponencial en l'ús de dispositius portàtils ha generat vastes quantitats de dades de sèries temporals, presentant oportunitats per a diverses aplicacions com el reconeixement d'activitats i el monitoratge de la salut. No obstant això, analitzar aquests conjunts de dades presenta reptes a causa de la seva complexitat i longitud. Per abordar això, aquesta tesi doctoral proposa la Representació de Canvis d'Estat per a Sèries Temporals (SCRTS, per les seves sigles en anglès), un mètode destinat a extreure informació rellevant relacionada amb la dinàmica de la sèrie temporal mentre es redueix significativament la dimensionalitat. A més, el SCRTS és independent de la longitud, la qual cosa permet l'aplicació d'aquest algoritme a marcs (valors consecutius de les variables en la sèrie temporal relacionada amb una classe determinada) de diferents longituds mentre es produeixen vectors de característiques del mateix tamany. Aquest aspecte és crucial per a les classificacions, assegurant uniformitat en les representacions de característiques per a marcs de diferent longitud. L'algoritme SCRTS es presenta en dues variants: l'enfocament unidimensional (1D-SCRTS) i el multidimensional (mD-SCRTS). En el 1D-SCRTS, cada marc està representat per una seqüència d'estats derivada de magnituds vectorials que resumeixen la informació de les variables interrelacionades en cada moment del temps de les mostres de la sèrie temporal. En contrast, el mD-SCRTS considera els valors de les variables de les mostres de forma individual abans de la discretització, el que li permet capturar informació relacionada amb tots els valors de les variables de la mostra de forma independent. L'efectivitat del SCRTS es demostra a través d'experiments de classificació d'activitats utilitzant tres conjunts de dades d'acceleròmetres. Tant el 1D-SCRTS com el mD-SCRTS exhibeixen capacitats sobresortints de reducció de dimensionalitat mentre aconsegueixen un rendiment de classificació considerable

ENG- The exponential growth in wearable device usage has generated vast amounts of time series data, presenting opportunities for various applications like activity recognition and health monitoring. However, analyzing these datasets poses challenges due to their complexity and length. To address this, this doctoral thesis proposes the State Changes Representation for Time Series (SCRTS), a method aimed at extracting relevant information related to the dynamics of the time series while significantly reducing the dimensionality. Moreover, SCRTS is length-independent, enabling the application of this algorithm to frames (consecutive values of the variables in the time series related to a given class) of varying lengths while producing feature vectors of the same size. This aspect is crucial for classifications, ensuring uniformity in feature representations across different time series lengths. The SCRTS algorithm is presented in two variants: the one-dimensional (1D-SCRTS) and the multidimensional (mD-SCRTS) approaches. In the 1D-SCRTS, each frame is represented by a sequence of states derived from vector magnitudes, which summarize the information of the interrelated variables at each time point of time series samples. In contrast, the mD-SCRTS considers individual variable values before discretization, allowing it to capture information related to all variable values independently. The effectiveness of the SCRTS is demonstrated through activity classification experiments using three accelerometer datasets. Both the 1D-SCRTS and the mD-SCRTS exhibit outstanding dimensionality reduction capabilities while achieving considerable classification performance

This work was carried out with the support of the Generalitat de Catalunya 2021 SGR 01125, and funded by the Grants for the Recruitment of New Research Staff (FI), provided by the Agència de Gestió d’Ajuts Universitaris i de Recerca (AGAUR)

Programa de Doctorat en Tecnologia

Keywords

Time series feature extraction, Característiques dels models Markov, Classificació de l'activitat, Clasificación de series temporales, Independent de la longitud, Length-independent, Visualización de características de series temporales, Activity classification, Acceleròmetres, Acelerómetros, Visualització de característiques de sèries temporals, Classificació de sèries temporals, Extracción de características de series temporales, Independiente de la longitud, Markov models features, 537, Time series features visualization, Clasificación de actividades, Time series classification, Accelerometers, Extracció de característiques de sèries temporals, Características de los modelos de Markov

  • BIP!
    Impact byBIP!
    selected citations
    These citations are derived from selected sources.
    This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    0
    popularity
    This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
    Average
    influence
    This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    Average
    impulse
    This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
    Average
Powered by OpenAIRE graph
Found an issue? Give us feedback
selected citations
These citations are derived from selected sources.
This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Citations provided by BIP!
popularity
This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
BIP!Popularity provided by BIP!
influence
This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Influence provided by BIP!
impulse
This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
BIP!Impulse provided by BIP!
0
Average
Average
Average
Green
Upload OA version
Are you the author of this publication? Upload your Open Access version to Zenodo!
It’s fast and easy, just two clicks!