Szczegóły publikacji
Opis bibliograficzny
Bronchopulmonary dysplasia prediction using Support Vector Machine and LIBSVM / Marcin OCHAB, Wiesław WAJS // W: FedCSIS [Dokument elektroniczny] : preprints of the Federated Conference on Computer Science and Information Systems : [Warsaw, Poland, 7 - 10 September, 2014] / PTI Polish Information Processing Society. — Wersja do Windows. — Dane tekstowe. — [Piscataway : IEEE], [2014]. — Dysk Flash. — e-ISBN: 978-83-60810-58-3. — S. 209–216. — Wymagania systemowe: Adobe Reader. — Bibliogr. s. 215–216, Abstr. — W bazie Web of Science: 2014 Federated Conference on Computer Science and Information Systems (FEDCSIS). — ISBN 978-83-60810-58-3. — S. 201–208
Autorzy (2)
Dane bibliometryczne
| ID BaDAP | 85381 |
|---|---|
| Data dodania do BaDAP | 2014-11-07 |
| DOI | 10.15439/2014F111 |
| Rok publikacji | 2014 |
| Typ publikacji | materiały konferencyjne (aut.) |
| Otwarty dostęp |  |
| Konferencja | Federated Conference on Computer Science and Information Systems |
Abstract
The paper presents BPD (Bronchopulmonary Dysplasia) prediction for extremely premature infants after their first week of life. SVM (Support Vector Machine) algorithm implemented in LIBSVMM[1] was used as classifier. Results are compared to others gathered in previous work [2] where LR (Logit Regression) and Matta) environment SVM implementation were used. Fourteen different risk factor parameters were considered and due to the high computational complexity only 3375 random combinations were analysed. Classifier based on eight feature model provides the highest accuracy which was 82.69%. The most promising 5-Nature model which gathered 82.23% was reasonably immune to random data changes and consistent with LR results. The main conclusion is that unlike Mat lab SVM[2] implementation, LIBSVM can be successfully used in considered problem, but it is less stable than LR. In addition, the article discusses influence of the model parameters selection on prediction quality.
