Diabetes prediction based on an improved whale optimization algorithm and support vector machine
Abstract
In recent years, the high incidence of diabetes and its complications has posed a significant threat to public health. To enhance the early diagnosis rate, this paper proposes a risk prediction model based on an improved Whale Optimization Algorithm (WOA) and Support Vector Machine (SVM). The model first uses a Gaussian kernel function to address the issue of nonlinear data being linearly separable in high-dimensional feature space. Then, it optimizes the Whale Optimization Algorithm by incorporating Tent mapping, opposition-based learning, nonlinear functions, and an adaptive inertia weight strategy. Subsequently, SVM parameters are optimized to determine the optimal penalty factor and kernel function parameters, improving the efficiency of SVM kernel parameter search. Finally, the optimized PAWOA-SVM model is applied to the diabetes dataset of Iraqi Medical Institutions and the Alibaba Cloud Tianchi Precision Medical Contest dataset for diabetes prediction and validation, achieving a prediction accuracy of 93.54%. The prediction results of the PAWOA-SVM model are compared with other commonly used models such as BP neural networks, PSO-SVM, and decision trees, demonstrating the superior predictive performance of the PAWOA-SVM model. Both the model’s recognition accuracy and computational efficiency show considerable improvement. Therefore, this research provides an effective model that can assist doctors in making early judgments about prediabetes, thereby improving the diagnosis rate of prediabetes.
References
1. Whiting DR, Guariguata L, Weil C, et al. IDF diabetes atlas: Global estimates of the prevalence of diabetes for 2011 and 2030. Diabetes research and clinical practice. 2011; 94(3): 311–321.
2. Jiang R, Xin Y, Chen Z, et al. A medical big data access control model based on fuzzy trust prediction and regression analysis. Applied Soft Computing. 2022; 117: 108423.
3. Jiang R, Han S, Yu Y, et al. An access control model for medical big data based on clustering and risk. Information Sciences. 2023; 621: 691–707.
4. Srivastava S, Sharma L, Sharma V, et al. Prediction of diabetes using artificial neural network approach. Springer Singapore Publishing; 2019. pp. 679–687.
5. Dudkina T, Meniailov I, Bazilevych K, et al. Classification and Prediction of Diabetes Disease using Decision Tree Method. IT&AS Publishing; 2021. pp. 163–172.
6. Akbulut FP, Akan A. Support vector machines combined with feature selection for diabetes diagnosis. IU-Journal of Electrical & Electronics Engineering. 2017; 17(1): 3257–3265.
7. Lin P. Research on prediction model of diabetes based on SVM. Jilin University; 2022.
8. Liu L, Zhang R. Multistrategy improved whale optimization algorithm and its application. Computational Intelligence and Neuroscience. 2022; 1: 3418269.
9. Veisi H. Introduction to SVM. In: Rad JA, Parand K, Chakraverty S (editors). Learning with Fractional Orthogonal Kernel Classifiers in Support Vector Machines: Theory, Algorithms and Applications. Springer Nature Singapore Publishing; 2023. pp. 3–18.
10. Azzeh M, Elsheikh Y, Nassif AB, et al. Examining the performance of kernel methods for software defect prediction based on support vector machine. Science of Computer Programming. 2023; 226: 102916.
11. Dai T, Dong Y. Introduction of SVM related theory and its application research. In: Proceedings of the 2020 3rd International Conference on Advanced Electronic Materials, Computers and Software Engineering (AEMCSE); 24–26 April 2020; Shenzhen, China.
12. Yao L, Fang Z, Xiao Y, et al. An intelligent fault diagnosis method for lithium battery systems based on grid search support vector machine. Energy. 2021; 214: 118866.
13. Resmini R, Silva L, Araujo AS, et al. Combining genetic algorithms and SVM for breast cancer diagnosis using infrared thermography. Sensors. 2021; 21(14): 4802.
14. Choubey DK, Tripathi S, Kumar P, et al. Classification of Diabetes by Kernel based SVM with PSO. Recent Advances in Computer Science and Communications. 2021; 14(4): 1242–1255.
15. Chakraborty S, Saha AK, Chakraborty R, et al. An enhanced whale optimization algorithm for large scale optimization problems. Knowledge-Based Systems. 2021; 233: 107543.
16. Godbole V, Gaikwad S. An Exponential Map-based Whale Optimization Algorithm (Exp-WOA) for Optimization. ECTI Transactions on Computer and Information Technology. 2024; 18(4): 443–455.
17. Wang X, Hu J, Hu J, et al. A modified equilibrium optimizer using opposition-based learning and teaching-learning strategy. IEEE Access. 2022; 10: 101408–101433.
18. Pan Z, Gu Z, Jiang X, et al. A modular approximation methodology for efficient fixed-point hardware implementation of the sigmoid function. IEEE Transactions on Industrial Electronics. 2022; 69(10): 10694–10703.
19. Zhang J, Wang JS. Improved whale optimization algorithm based on nonlinear adaptive weight and golden sine operator. IEEE Access. 2020; 8: 77013–77048.
20. Zhao X, Fang Y, Liu L, et al. A covariance-based Moth–flame optimization algorithm with Cauchy mutation for solving numerical optimization problems. Applied Soft Computing. 2022; 119: 108538.
21. Singh D, Singh B. Investigating the impact of data normalization on classification performance. Applied Soft Computing. 2020; 97: 105524.
Copyright (c) 2024 Xinyi Yang, Jingyi Yang, Xiaoyan Liu, Qiang Hu
This work is licensed under a Creative Commons Attribution 4.0 International License.
Submit a Paper
