Parkinson’s disease is a neurodegenerative disorder and afects the nerve cells that produce dopamine in the brain. In this paper, we investigated comparative studies on the diferent scenarios such as AutoEncoder and Ant Colony Optimization feature selection algorithms for the efective features in diagnosis of Parkinson’s disease. These algorithms are implemented to the voice dataset obtained from online repository. Then selected features are presented to the Decision tree, SVM, K-NN, Ensemble, Naive Bayes and Discriminant classifers for each of the binary classifcation problems. The proposed methods are evaluated with the sensitivity, specifcity, precision, recall and accuracy criteria. The proposed systems are trained and tested with these classifers separately to carry out a comparative study and to analyse the success of feature selection methods in discriminating healthy people and PD patients. In Parkinson’s data there are 24 features that obtained from the signal voices. Some of the features in training of the classifer have problems and these problems reduce the accuracy of the system. It is found that for K-NN and Ensemble classifcation methods both ACO and Autoencoder have the same and the best training performance. Testing results show that accuracy rate of ACO is higher than Autoencoder method.
Eser Adı (dc.title) | An Approach Using in Communication Network Apply in Healthcare System Based on the Deep Learning Autoencoder Classifcation Optimization Metaheuristic Method |
Yazar (dc.contributor.author) | Cevat Rahebi |
Yayın Yılı (dc.date.issued) | 2023 |
Tür (dc.type) | Makale |
Özet (dc.description.abstract) | Parkinson’s disease is a neurodegenerative disorder and afects the nerve cells that produce dopamine in the brain. In this paper, we investigated comparative studies on the diferent scenarios such as AutoEncoder and Ant Colony Optimization feature selection algorithms for the efective features in diagnosis of Parkinson’s disease. These algorithms are implemented to the voice dataset obtained from online repository. Then selected features are presented to the Decision tree, SVM, K-NN, Ensemble, Naive Bayes and Discriminant classifers for each of the binary classifcation problems. The proposed methods are evaluated with the sensitivity, specifcity, precision, recall and accuracy criteria. The proposed systems are trained and tested with these classifers separately to carry out a comparative study and to analyse the success of feature selection methods in discriminating healthy people and PD patients. In Parkinson’s data there are 24 features that obtained from the signal voices. Some of the features in training of the classifer have problems and these problems reduce the accuracy of the system. It is found that for K-NN and Ensemble classifcation methods both ACO and Autoencoder have the same and the best training performance. Testing results show that accuracy rate of ACO is higher than Autoencoder method. |
Açık Erişim Tarihi (dc.date.available) | 2023-11-25 |
Yayıncı (dc.publisher) | Wireless Personal Communications |
Dil (dc.language.iso) | En |
Konu Başlıkları (dc.subject) | Ant colony optimization |
Konu Başlıkları (dc.subject) | Autoencoder deep learning |
Konu Başlıkları (dc.subject) | Feature selection |
Konu Başlıkları (dc.subject) | Parkinson’s disease |
Tek Biçim Adres (dc.identifier.uri) | https://hdl.handle.net/20.500.14081/1918 |
Dergi (dc.relation.journal) | Wireless Personal Communications |
DOI (dc.identifier.doi) | 10.1007/s11277-023-10759-9 |
Orcid (dc.identifier.orcid) | 0000-0001-9875-4860 |