Comparison of Two Types of Neural Networks for Classifying the Effects of Heart Diseases in Erbil City
DOI:
https://doi.org/10.21271/zjhs.29.6.13Keywords:
Classification, Neural Network, Radial Basic Function , Multilayer Perceptron Network and algorithmsAbstract
The main objective of this research is to compare two types of neural networks: the Radial Basis Function (RBF) network and the Multilayer Perceptron (MLP) network, to determine which provides better performance in classifying heart disease. The data were collected at the Surgical Specialty Hospital – Cardiac Center in Erbil City, comprising 196 observations recorded between January 1st, 2006, and April 30th, 2023. To evaluate model performance, the study employed common classification metrics, including Mean Square Error (MSE) and Mean Absolute Percentage Error (MAPE). Both neural network models were applied to the dataset. The RBF network demonstrated superior performance, achieving the lowest error values in both training and testing phases, while also requiring less training time. Therefore, the RBF network can be considered a reliable and highly effective method for classifying heart disease cases in Erbil City.
References
- Adelson, S. & Keys, A. (1962). Diet and health characteristics of 123 business and professional men and methods used to obtain the dietary information. ARS, US Department of Agriculture, 11-62.
- Bergson, A.W. & Choatel, T. (1996) ‘Cooperative-Competitive Genetic Evolution of Radial Basis Function Centers and Widths for Time Series Prediction’, IEEE Transactions on Neural Networks, 7(3), pp. 869–880.
- Bianchini, M. & Scarselli, F. (2014) ‘On the Complexity of Neural Network Classifiers: A Comparison between Shallow and Deep Architectures’, IEEE Transactions on Eurasian J. Sci. Eng., 25(8), pp. 1553–1565.
- Bogdan, M. (2013) ‘Comparison of Training Algorithms and Network Architectures’, 17th IEEE Intelligent Engineering Systems Conference, Costa Rica, p. 11.
- Bors, A. & Gabboui, G. (1994) ‘Minimal topology for a Radial Basis Function Neural Network for Pattern Classification’, Digital Signal Processing: A Review Journal, 4(3), pp. 173–188.
- Dan, S. (2002) ‘Training Radial Basis Neural Networks with the Extended Kalman Filter’, Neurocomputing, 48, pp. 455–475.
- Gershenson, C. (1998) Artificial Neural Networks for Beginners. Sussex Academy, UK.
- Graupe, D. (2007) Principles of Artificial Neural Networks. 2nd ed. World Scientific Publishing Co., Singapore; London.
- Hagan, M., Demuth, H. & Beale, M. (1996) Neural Network Design. PWS Publishing Co., USA.
- Hao Yu, Philip, Tiantian, X., Tonasz, B. & Bogdan, W. (2014) ‘An Incremental Design of Compact Radial Basis Function Networks’, IEEE Transactions on Neural Networks and Learning Systems (accepted).
- Hassan, S. (2022) A New Self-Organizing Map (SOM) to Forecast the Electricity Generation in KR-Iraq. PhD thesis, College of Administration and Economics, Salahaddin University-Erbil.
- Haykin, S. (1999) Neural Networks: A Comprehensive Foundation. 2nd ed. Prentice-Hall, MC Master University, Ontario, Canada.
- Nicolas, B. (1999) ‘Reformulated Radial Basis Neural Networks Trained by Gradient Descent’, IEEE Transactions on Neural Networks, 10(3), pp. 657–671.
- Principe, J., Euliano, N. & Lefebvre, W. (2000) Neural and Adaptive Systems: Fundamentals through Simulation. John Wiley & Sons, New York.
- Saalih, W. & Hassan, S. (2023) ‘Using Neural Networks to Forecast the Electricity Generation in Kurdistan Region-Iraq’, 7th International Conference of Union of Arab Statisticians, Cairo, Egypt.
- Sinha, H. (2002) Designing a Neural Network for Forecasting Financial and Economic Time Series. June 4th, IND E 593.
- Zurada, M. & Cholew, J. (1994) ‘Introduction: Artificial Neural Systems’, Proceedings of the IEEE International Joint Conference on Neural Networks, Houston, Texas, USA.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 Samyia Khalid Hasan
This work is licensed under a Creative Commons Attribution 4.0 International License.
