Improving stability of feature selection methods
The result's identifiers
Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F07%3A03134587" target="_blank" >RIV/68407700:21230/07:03134587 - isvavai.cz</a>
Result on the web
—
DOI - Digital Object Identifier
—
Alternative languages
Result language
angličtina
Original language name
Improving stability of feature selection methods
Original language description
An improper design of feature selection methods can often lead to incorrect conclusions. Moreover, it is not generally realised that functional values of the criterion guiding the search for the best feature set are random variables with some probabilitydistribution. This contribution examines the influence of several estimation techniques on the consistency of the final result. We propose an entropy based measure which can assess the stability of feature selection methods with respect to perturbationsin the data. Results show that filters achieve a better stability and performance if more samples are employed for the estimation, i.e., using leave-one-out cross-validation, for instance. However, the best results for wrappers are acquired with the 50/50 holdout validation.
Czech name
Improving stability of feature selection methods
Czech description
An improper design of feature selection methods can often lead to incorrect conclusions. Moreover, it is not generally realised that functional values of the criterion guiding the search for the best feature set are random variables with some probabilitydistribution. This contribution examines the influence of several estimation techniques on the consistency of the final result. We propose an entropy based measure which can assess the stability of feature selection methods with respect to perturbationsin the data. Results show that filters achieve a better stability and performance if more samples are employed for the estimation, i.e., using leave-one-out cross-validation, for instance. However, the best results for wrappers are acquired with the 50/50 holdout validation.
Classification
Type
D - Article in proceedings
CEP classification
JD - Use of computers, robotics and its application
OECD FORD branch
—
Result continuities
Project
<a href="/en/project/1M0567" target="_blank" >1M0567: Centre for Applied Cybernetics</a><br>
Continuities
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)
Others
Publication year
2007
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Data specific for result type
Article name in the collection
CAIP 2007: Proceedings of the 12th International Conference on Computer Analysis of Images and Patterns
ISBN
978-3-540-74271-5
ISSN
—
e-ISSN
—
Number of pages
8
Pages from-to
929-936
Publisher name
Springer
Place of publication
Berlin
Event location
Vienna
Event date
Aug 27, 2007
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
—