On Coupling Robust Estimation with Regularization for High-Dimensional Data

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F67985807%3A_____%2F17%3A00474072" target="_blank" >RIV/67985807:_____/17:00474072 - isvavai.cz</a>

Výsledek na webu

<a href="http://dx.doi.org/10.1007/978-3-319-55723-6_2" target="_blank" >http://dx.doi.org/10.1007/978-3-319-55723-6_2</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1007/978-3-319-55723-6_2" target="_blank" >10.1007/978-3-319-55723-6_2</a>

Jazyk výsledku

angličtina

Název v původním jazyce

On Coupling Robust Estimation with Regularization for High-Dimensional Data

Popis výsledku v původním jazyce

Standard data mining procedures are sensitive to the presence of outlying measurements in the data. Therefore, robust data mining procedures are highly desirable, which are resistant to outliers. This work has the aim to propose new robust classification procedures for high-dimensional data and algorithms for their efficient computation. Particularly, we use the idea of implicit weights assigned to individual observation to propose several robust regularized versions of linear discriminant analysis (LDA), suitable for data with the number of variables exceeding the number of observations. The approach is based on a regularized version of the minimum weighted covariance determinant (MWCD) estimator and represents a unique attempt to combine regularization and high robustness, allowing to down-weight outlying observations. Classification performance of new methods is illustrated on real fMRI data acquired in neuroscience research.

Název v anglickém jazyce

On Coupling Robust Estimation with Regularization for High-Dimensional Data

Popis výsledku anglicky

Standard data mining procedures are sensitive to the presence of outlying measurements in the data. Therefore, robust data mining procedures are highly desirable, which are resistant to outliers. This work has the aim to propose new robust classification procedures for high-dimensional data and algorithms for their efficient computation. Particularly, we use the idea of implicit weights assigned to individual observation to propose several robust regularized versions of linear discriminant analysis (LDA), suitable for data with the number of variables exceeding the number of observations. The approach is based on a regularized version of the minimum weighted covariance determinant (MWCD) estimator and represents a unique attempt to combine regularization and high robustness, allowing to down-weight outlying observations. Classification performance of new methods is illustrated on real fMRI data acquired in neuroscience research.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

<a href="/cs/project/GA13-23940S" target="_blank" >GA13-23940S: Osobnost a spontánní mozková aktivita během klidu a sledování filmu: vzájemný vztah a strukturní determinanty</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2017

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Data Science. Innovative Developments in Data Analysis and Clustering

ISBN

978-3-319-55722-9

ISSN

1431-8814

e-ISSN

—

Počet stran výsledku

13

Strana od-do

15-27

Název nakladatele

Springer

Místo vydání

Cham

Místo konání akce

Bologna

Datum konání akce

5. 7. 2015

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—