Number of the records: 1
A dimension reduction in neural network using copula matrix
- 1.
SYSNO ASEP 0561617 Document Type J - Journal Article R&D Document Type Journal Article Subsidiary J Článek ve WOS Title A dimension reduction in neural network using copula matrix Author(s) Sheikhi, A. (IR)
Mesiar, R. (CZ)
Holeňa, Martin (UIVT-O) SAI, RIDNumber of authors 3 Source Title International Journal of General Systems. - : Taylor & Francis - ISSN 0308-1079
Roč. 52, č. 2 (2023), s. 131-146Number of pages 16 s. Publication form Print - P Language eng - English Country GB - United Kingdom Keywords Principal component ; copula ; neural network ; correlation ; association measure OECD category Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8) Method of publishing Limited access Institutional support UIVT-O - RVO:67985807 UT WOS 000846787700001 EID SCOPUS 85136843132 DOI 10.1080/03081079.2022.2108029 Annotation In prediction analysis, there may exist some nonlinear relations between the exploratory variables, which are not captured by traditional correlation-based linear models such as multiple regression, principal component regression, and so on. In this work, we employ a copula matrix to extract principal components of a set of variables which are pair-wisely associated with a copula. By estimating the pairwise copula and its corresponding parameter(s), we suggest an optimization method to extract principal components from a matrix which contains some pairwise measures of association. We use these components as inputs of an artificial neural network to make a more accurate prediction. We test our proposed method using a simulation study and use it to carry out a more accurate prediction in an AIDS as well as a COVID-19 dataset. To increase the reliability of results, we employ a cross-validation technique. Workplace Institute of Computer Science Contact Tereza Šírová, sirova@cs.cas.cz, Tel.: 266 053 800 Year of Publishing 2024 Electronic address https://dx.doi.org/10.1080/03081079.2022.2108029
Number of the records: 1