Performance Comparison of Support Vector Machines,Random Forest and Artificial Neural Networks inBinary Classification: Descriptive Comparison Study
| dc.contributor.author | Akkuş, Zeki | |
| dc.contributor.author | Dırıcan, Emre | |
| dc.date.accessioned | 2024-04-24T19:13:07Z | |
| dc.date.available | 2024-04-24T19:13:07Z | |
| dc.date.issued | 2021 | |
| dc.department | Dicle Üniversitesi | en_US |
| dc.description.abstract | Objective: In this study, it was aimed to find themethod with high classification success among the methods used inthe study by comparing the supervised machine learning methodsaccording to the classification performance. Material and Meth ods: In our study, both the real data set obtained from 302 patientswith invasive ductal carcinoma and 24 different data sets obtainedby simulation were used to compare the classification performanceof support vector machines, random forest and artificial neural net works. The success of classifications of the methods used wascompared according to the general accuracy, F-measure, Matthewscorrelation coefficient, area under the curve (AUC) and discrimi nant power in breast cancer data. In addition, the difference in train ing-test accuracy in the simulation data and the significance of thisdifference were also evaluated. Results: The highest survival classifi cation accuracy (80%) for the test set of stage III patients with inva sive ductal carcinoma was obtained from support vector machines(SVM) with the radial-based kernel. The highest values in other per formance metrics (F-measure=0.87, Matthews correlation coeffi cient=0.22, AUC=0.89 and discriminant power=0.52), and the mostsuccessful results in simulation data were generally obtained fromSVM. Conclusion: SVM had higher accuracy in both the real data setand simulation data than random forest and artificial neural networks. | en_US |
| dc.identifier.doi | 10.5336/biostatic.2021-81105 | |
| dc.identifier.endpage | 251 | en_US |
| dc.identifier.issn | 1308-7894 | |
| dc.identifier.issn | 2146-8877 | |
| dc.identifier.issue | 3 | en_US |
| dc.identifier.startpage | 236 | en_US |
| dc.identifier.trdizinid | 503915 | |
| dc.identifier.uri | https://doi.org/10.5336/biostatic.2021-81105 | |
| dc.identifier.uri | https://search.trdizin.gov.tr/yayin/detay/503915 | |
| dc.identifier.uri | https://hdl.handle.net/11468/28380 | |
| dc.identifier.volume | 13 | en_US |
| dc.indekslendigikaynak | TR-Dizin | |
| dc.language.iso | en | en_US |
| dc.relation.ispartof | Türkiye Klinikleri Biyoistatistik Dergisi | |
| dc.relation.publicationcategory | Makale - Ulusal Hakemli Dergi - Kurum Öğretim Elemanı | en_US |
| dc.rights | info:eu-repo/semantics/openAccess | en_US |
| dc.title | Performance Comparison of Support Vector Machines,Random Forest and Artificial Neural Networks inBinary Classification: Descriptive Comparison Study | en_US |
| dc.title | Performance Comparison of Support Vector Machines,Random Forest and Artificial Neural Networks inBinary Classification: Descriptive Comparison Study | |
| dc.type | Article | en_US |
