ScienceDirect Available online at

---

ScienceDirect
Available online at 
www.sciencedirect.com
Procedia Computer Science 120 (2017) 588–593
1877-0509 
©
2018 The Authors. Published by Elsevier B.V.
Peer-review under responsibility of the scientific committee of the 9th International Conference on Theory and application of Soft 
Computing, Computing with Words and Perception.
10.1016/j.procs.2017.11.283
9th International Conference on Theory and Application of Soft Computing, Computing with 
Words and Perception, ICSCCW 2017, 24-25 August 2017, Budapest, Hungary
10.1016/j.procs.2017.11.283
1877-0509
Available online at 
www.sciencedirect.com
ScienceDirect
Procedia Computer Science 00 (2018) 000–000 
www.elsevier.com/locate/procedia 
1877-0509© 2018 The Authors. Published by Elsevier B.V. 
Peer-review under responsibility of the scientific committee of the 9th International Conference on Theory and application of 
Soft Computing, Computing with Words and Perception.
9th International Conference on Theory and Application of Soft Computing, Computing with 
Words and Perception, ICSCCW 2017, 22-23 August 2017, Budapest, Hungary 
Diagnosis of heart disease using genetic algorithm based trained 
recurrent fuzzy neural networks 
Kaan Uyar
a
*, Ahmet 
İ
lhan
a
a
Near East University, P.O.BOX: 99138, Nicosia, TRNC, Mersin 10, Turkey
Abstract 
The World Health Organization (WHO) estimated one third of all global deaths reason as cardiovascular diseases in 2015. Some 
computational techniques were proposed for investigation of heart diseases. This study proposes a genetic algorithm (GA) based 
trained recurrent fuzzy neural networks (RFNN) to diagnosis of heart diseases. The University of California Irvine (UCI) 
Cleveland heart disease dataset is used in this study. Out of total 297 instances of patient data, 252 are used for training and 45 of 
them are chosen to be the testing. The results showed that 97.78% accuracy was obtained from testing set. In addition to the 
accuracy, root mean square error, the probability of the misclassification error, specificity, sensitivity, precision and F-score are 
calculated. The results were found to be satisfying based on comparison. 
© 2018 The Authors. Published by Elsevier B.V. 
Peer-review under responsibility of the scientific committee of the 9th International Conference on Theory and application of 
Soft Computing, Computing with Words and Perception. 
Keywords:
Heart disease; recurrent fuzzy neural networks.