Fast Dependency-Aware Feature Selection in Very-High-Dimensional Pattern Recognition

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F67985556%3A_____%2F11%3A00365937" target="_blank" >RIV/67985556:_____/11:00365937 - isvavai.cz</a>

Výsledek na webu

—

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Fast Dependency-Aware Feature Selection in Very-High-Dimensional Pattern Recognition

Popis výsledku v původním jazyce

The paper addresses the problem of making dependency-aware feature selection feasible in pattern recognition problems of very high dimensionality. The idea of individually best ranking is generalized to evaluate the contextual quality of each feature ina series of randomly generated feature subsets. Each random subset is evaluated by a criterion function of arbitrary choice (permitting functions of high complexity). Eventually, the novel dependency-aware feature rank is computed, expressing the averagebenefit of including a feature into feature subsets. The method is efficient and generalizes well especially in very-high-dimensional problems, where traditional context-aware feature selection methods fail due to prohibitive computational complexity orto over-fitting. The method is shown well capable of over-performing the commonly applied individual ranking which ignores important contextual information contained in data.

Název v anglickém jazyce

Fast Dependency-Aware Feature Selection in Very-High-Dimensional Pattern Recognition

Popis výsledku anglicky

The paper addresses the problem of making dependency-aware feature selection feasible in pattern recognition problems of very high dimensionality. The idea of individually best ranking is generalized to evaluate the contextual quality of each feature ina series of randomly generated feature subsets. Each random subset is evaluated by a criterion function of arbitrary choice (permitting functions of high complexity). Eventually, the novel dependency-aware feature rank is computed, expressing the averagebenefit of including a feature into feature subsets. The method is efficient and generalizes well especially in very-high-dimensional problems, where traditional context-aware feature selection methods fail due to prohibitive computational complexity orto over-fitting. The method is shown well capable of over-performing the commonly applied individual ranking which ignores important contextual information contained in data.

Druh

D - Stať ve sborníku

CEP obor

IN - Informatika

OECD FORD obor

—

Projekt

<a href="/cs/project/1M0572" target="_blank" >1M0572: Data, algoritmy, rozhodování</a><br>

Návaznosti

Z - Vyzkumny zamer (s odkazem do CEZ)

Rok uplatnění

2011

Kód důvěrnosti údajů

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

Název statě ve sborníku

Proceedings of the IEEE International Conference on Systems, Man, and Cybernetics (IEEE SMC 2011)

ISBN

978-1-4577-0653-0

ISSN

—

e-ISSN

—

Počet stran výsledku

8

Strana od-do

502-509

Název nakladatele

IEEE

Místo vydání

Piscataway

Místo konání akce

Anchorage, Alaska

Datum konání akce

9. 10. 2011

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—