Evaluation of The Classification Performance of Methods Developed in ANN Training
Yükleniyor...
Dosyalar
Tarih
2021
Yazarlar
Dergi Başlığı
Dergi ISSN
Cilt Başlığı
Yayıncı
Selçuk Üniversitesi
Erişim Hakkı
info:eu-repo/semantics/openAccess
Özet
In this study, the performance and success of some models used in the training of artificial neural networks are compared. The UCI machine learning database was used for the performance evaluation and comparison process and for the IRIS dataset), which we often encounter in performance evaluations. In the study, the classification performances of our data sets were calculated using Feedforward Neural Network Cross-Validation (FNN-CV), Probabilistic Neural Network Cross-Validation (PNN-CV) and Recurrent Neural Network Cross-Validation (RNNCV) learning techniques. Our data has been classified using a 10-fold cross validation technique. The RNN-CV method for the IRIS data set proved to be a good classification method by achieving 100% accuracy success which shows that it should be used in classification problems.
Açıklama
Anahtar Kelimeler
Artificial Neural Network, Cross-Validation, Feedforward Neural Network, Probabilistic Neural Network, Recurrent Neural Network
Kaynak
Selcuk University Journal of Engineering Sciences
WoS Q Değeri
Scopus Q Değeri
Cilt
20
Sayı
1
Künye
Yasar, A., Saritas, I., (2021). Evaluation of The Classification Performance of Methods Developed in ANN Training. Selcuk University Journal of Engineering Sciences, 20 (01), 1-6.
