Počet záznamů: 1
Effective Automatic Method Selection for Nonlinear Regression Modeling
- 1.
SYSNO ASEP 0541777 Druh ASEP J - Článek v odborném periodiku Zařazení RIV J - Článek v odborném periodiku Poddruh J Článek ve WOS Název Effective Automatic Method Selection for Nonlinear Regression Modeling Tvůrce(i) Kalina, Jan (UIVT-O) RID, SAI, ORCID
Neoral, Aleš (UIVT-O) RID, SAI
Vidnerová, Petra (UIVT-O) RID, SAI, ORCIDČíslo článku 2150020 Zdroj.dok. International Journal of Neural Systems. - : World Scientific Publishing - ISSN 0129-0657
Roč. 31, č. 10 (2021)Poč.str. 12 s. Forma vydání Online - E Jazyk dok. eng - angličtina Země vyd. SG - Singapur Klíč. slova metalearning ; nonlinear regression ; robust statistical estimation ; feature selection ; AutoML Vědní obor RIV IN - Informatika Obor OECD Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8) CEP GA19-05704S GA ČR - Grantová agentura ČR GA18-23827S GA ČR - Grantová agentura ČR Způsob publikování Omezený přístup Institucionální podpora UIVT-O - RVO:67985807 UT WOS 000696596800003 EID SCOPUS 85104028019 DOI 10.1142/S0129065721500209 Anotace Metalearning, an important part of artificial intelligence, represents a promising approach for the task of automatic selection of appropriate methods or algorithms. This paper is interested in recommending a suitable estimator for nonlinear regression modeling, particularly in recommending either the standard nonlinear least squares estimator or one of such available alternative estimators, which is highly robust with respect to the presence of outliers in the data. The authors hold the opinion that theoretical considerations will never be able to formulate such recommendations for the nonlinear regression context. Instead, metalearning is explored here as an original approach suitable for this task. In this paper, four different approaches for automatic method selection for nonlinear regression are proposed and computations over a training database of 643 real publicly available datasets are performed. Particularly, while the metalearning results may be harmed by the imbalanced number of groups, an effective approach yields much improved results, performing a novel combination of supervised feature selection by random forest and oversampling by synthetic minority oversampling technique (SMOTE). As a by-product, the computations bring arguments in favor of the very recent nonlinear least weighted squares estimator, which turns out to outperform other (and much more renowned) estimators in a quite large percentage of datasets. Pracoviště Ústav informatiky Kontakt Tereza Šírová, sirova@cs.cas.cz, Tel.: 266 053 800 Rok sběru 2022 Elektronická adresa http://dx.doi.org/10.1142/S0129065721500209
Počet záznamů: 1