A Comparative Study for Enhancing Classification Accuracy Using (SVM) Models: An Applied Study
Keywords:
Binary-response Support Vector Machine, Multi-response Support Vector Machine, Radial Basis Function (RBF), Polynomial Function, Sigmoid FunctionAbstract
This study aimed to evaluate the performance of Support Vector Machine (SVM) models in data classification, examining the effect of the type of kernel function and the nature of the response on classification accuracy, using the statistical software ORANGE. Three different performance evaluation methods were applied: Test On Train Data, Cross Validation, and Random Sampling. The results indicated that classification accuracy in SVM models varies depending on the kernel function used and the response type. For binary-response models, the Radial Basis Function (RBF) achieved the highest classification accuracy compared to Polynomial and Sigmoid functions, with the evaluation methods ranked as follows: Test On Train Data (highest), followed by Cross Validation, and Random Sampling (lowest). In multi-response models, the highest accuracy was obtained using the Polynomial function compared to RBF and Sigmoid, maintaining the same ranking of evaluation methods.
Based on these findings, the study recommends adopting multi-response SVM models in future statistical studies involving multi-response data due to their high classification accuracy. It also emphasizes the use of a confusion matrix as an effective tool to evaluate model performance by comparing correctly and incorrectly classified positive and negative cases. Additionally, the study highlights the importance of enhancing healthcare services for heart disease patients by providing specialized hospitals and medical centers and ensuring access to necessary treatment, alongside supporting and encouraging research and studies related to cardiovascular diseases to improve therapeutic outcomes and deepen understanding of the condition.
References
جاب الله, & محمود أسامة مصطفى السيد. (2024). استخدام تقنية SVM في رصد وتقييم دور الإنسان كعامل جيومورفولوجي على سواحل بحيرة المنزلة خلال الفترة الممتدة بين عامي (1984م-2024م) دراسة في الجيومورفولوجيا التطبيقية. المجلة العلمية بکلية الآداب, 2024(56), 708-736.
سالم، أحمد محمد. (2025). المقارنة بين نماذج آلة المتجه الداعم في تصنيف البيانات (دراسة تطبيقية). مجلة البحوث المالية والتجارية، 26(3)، 138–162.
Amari, S., & Wu, S. (2021). Improving kernel-based classification models for biomedical data. Journal of Machine Learning Research, 22(1), 1–25.
Ben-Hur, A., & Weston, J. (2019). A user’s guide to support vector machines. In Methods in Molecular Biology.
Cai, W., Cai, M., Li, Q., & Liu, Q. (2023). Three way Imbalanced Learning based on Fuzzy Twin SVM. arXiv. https://arxiv.org/abs/2301.
Chen, L., Zhang, Y., & Wu, X. (2023). Machine learning-based heart disease prediction using clinical and demographic features. IEEE Access.
Duarte, J., & Ribeiro, J. (2023). Advances in multi-class support vector machines for educational data classification. International Journal of Data Science and Analytics, 17(1), 25–41.
El Maouaki, W., Said, T., & Bennai, M. (2024). Quantum support vector machine for prostate cancer detection: A performance analysis. arXiv. https://arxiv.org/
Halagundegowda, G. R., Singh, A., T. L., M. K., & Naveena, K. (2023). Evaluation of classification ability of support vector machine (SVM) in binary classification problems. International Journal of Statistical Applications and Mathematics, 8(5), 07–13.
Hastie, T., Tibshirani, R., & Friedman, J. (2022). The Elements of Statistical Learning: Data Mining, Inference, and Prediction (2nd ed.). Springer.
Hsu, C.-W., & Lin, C.-J. (2021). A Comparison of Methods for Multi-class Support Vector Machines. IEEE Transactions on Neural Networks, 32(6), 2345–2357.
Islam, M., Rahman, S., & Akter, N. (2024). Advanced predictive modeling for heart disease using hybrid machine learning techniques. Expert Systems with Applications.
Karam, F., & Chen, D. (2024). Socioeconomic predictors of student achievement: A global perspective. International Journal of Educational Research, 125.
Khan, T., Alshurideh, M., & Salloum, S. (2023). Predicting student performance using machine learning: A systematic review. Education Sciences, 13(4).
Kowsari, K., Brown, D., & Barnes, L. (2022). Evaluation of kernel-based support vector machines for multi-class problems. Machine Learning and Data Analytics Review, 12(2), 98–115.
Kumar, A., & Mishra, D. (2025). Improving Support Vector Machine using Modified Kernel Function. International Journal of Scientific Research and Modern Technology, 4(5), 1–5.
Lee, J., & Kim, S. (2024). Gender differences in academic performance among secondary students. Journal of Educational Psychology, 116(2).
Li, Q., & Wang, P. (2024). Cardiovascular risk prediction using optimized classification algorithms. Journal of Biomedical Informatics.
Müller, K. R., Mika, S., Rätsch, G., Tsuda, K., & Schölkopf, B. (2022). An introduction to kernel-based learning algorithms. IEEE Transactions on Neural Networks and Learning Systems.
Okfalisa, S., Ahmed, A., et al. (2021). SVM is a classification method that produces a fairly high degree of accuracy. Journal of Engineering Science and Technology, 16(2), 1372–138?. JESTEC.
Powers, D. M. W. (2020). Evaluation: From Precision, Recall and F-Measure to ROC, Informedness, Markedness & Correlation. Journal of Machine Learning Technologies, 2(1), 37–63.
Rahman, M., & Yadav, R. (2023). Heart disease prediction using supervised learning models: A comparative study. Applied Intelligence.
Roy, A., & Singh, R. (2024). Multiclass classification approaches for predicting students’ academic outcomes. Applied Intelligence, 54.
Sahu, S. K., & Tripathy, B. K. (2023). Applications, Strengths and Limitations of Support Vector Machines in Data Classification. Journal of Machine Learning Research, 24(12), 1–19.
Schölkopf, B., & Smola, A. J. (2018). Learning with kernels: Support vector machines, regularization, optimization, and beyond. MIT Press.
Sholihah, W., Handayani, A. S., & Sarjana, S. (2023). Revolutionizing healthcare: Comprehensive evaluation and optimization of SVM kernels for precise general health diagnosis. Scientific Journal of Informatics, 10(4). https://journal.unnes.ac.id
Zhang, L., & Wu, X. (2023). An Improved Gaussian Kernel for SVM Classification. Journal of Intelligent & Fuzzy Systems, 45(2), 1957–1972.
Zhang, Y., & Wang, S. (2021). Support vector machines for multi-class classification: A comparative study of kernel functions. Journal of Intelligent Systems, 30(4), 543–558.
Zhu, L., Chen, H., & Li, Y. (2023). Age-related factors affecting academic performance: A meta-analysis. Learning and Instruction, 85.
