Szczegóły publikacji
Opis bibliograficzny
Effect of feature discretization on classification performance of explainable scoring-based machine learning model / Arkadiusz Pajor, Jakub Żołnierek, Bartłomiej ŚNIEŻYŃSKI, Arkadiusz Sitek // W: Computational Science – ICCS 2022 : 22nd international conference : London, UK, June 21–23, 2022 : proceedings, Pt. 3 / eds. Derek Groen, Clélia de Mulatier, Maciej Paszyński, Valeria V. Krzhizhanovskaya, Jack J. Dongarra, Peter M. A. Sloot. — Cham : Springer Nature Switzerland, cop. 2022. — (Lecture Notes in Computer Science ; ISSN 0302-9743 ; LNCS 13352). — ISBN: 978-3-031-08756-1; e-ISBN: 978-3-031-08757-8. — S. 92–105. — Bibliogr., Abstr. — Publikacja dostępna online od: 2022-06-15. — A. Pajor - pierwsza afiliacja: Sano Centre for Computational Medicine, Cracow
Autorzy (4)
- AGHPajor Arkadiusz
- Żołnierek Jakub
- AGHŚnieżyński Bartłomiej
- Sitek Arkadiusz
Dane bibliometryczne
| ID BaDAP | 140673 |
|---|---|
| Data dodania do BaDAP | 2023-01-11 |
| DOI | 10.1007/978-3-031-08757-8_9 |
| Rok publikacji | 2022 |
| Typ publikacji | materiały konferencyjne (aut.) |
| Otwarty dostęp |  |
| Wydawca | Springer |
| Konferencja | International Conference on Computational Science 2022 |
| Czasopismo/seria | Lecture Notes in Computer Science |
Abstract
We improve the utility of the Risk-calibrated Supersparse Linear Integer Model (RiskSLIM). It is a scoring system that is an interpretable machine learning classification model optimized for performance. Scoring systems are commonly used in healthcare and justice. We implement feature discretization (FD) in the hyperparameter optimization process to improve classification performance and refer to the new approach as FD-RiskSLIM. We test the approach using two medical applications. We compare the results of FD-RiskSLIM, RiskSLIM, and other machine learning (ML) models. We demonstrate that scoring models based on RiskSLIM, in addition to being interpretable, perform at least on par with the state-of-the-art ML models such as Gradient Boosting in terms of classification metrics. We show the superiority of FD-RiskSLIM over RiskSLIM.
