Density based fuzzy support vector machine: application to diabetes dataset

In this work, we propose a deep prediction diabetes system based on a new version of the support vector machine optimization model.  First, we determine three types of patients (noisy, cord, and interior) basing on specific parameters. Second, we equilibrate the clinical data sets by suppressing noisy and cord patients.  Third, we determine the support vectors by solving an optimization program with a reasonable size. Our system is performed on the well-known diabetes dataset PIMA.  The experimental results show that the proposed method improves the prediction accuracy and the proposed system significantly outperforms all other versions of SVM as well as literature methods of classification.

опорна векторна машина
машинне навчання
оптимізація
класифікація
diabetes
  1. WHO. Diabetes.  Available online: https://www.who.int/news-room/fact-sheets/detail/diabetes (accessed on 1 June  2020).
  2. IDF Diabetes Atlas, A.D. Type 2 Diabetes.  Available online: https://www.idf.org/aboutdiabetes/type-2-diabetes.html (accessed on 20 March 2020).
  3. El Moutaouakil K., Touhafi A.  A New Recurrent Neural Network Fuzzy Mean Square Clustering Method.  2020 5th International Conference on Cloud Computing and Artificial Intelligence: Technologies and Applications (CloudTech). 1–5 (2020).
  4. Vapnik V. N.  The Nature of Statistical Learning Theory.  Springer Science and Business Media (1999).
  5. Burges C. J. C.  A tutorial on support vector machines for pattern recognition.  Data Mining and Knowledge Discovery. 2, 121–167 (1998).
  6. Vapnik V. N.,  Chervonenkis A. Ya.  A class of algorithms for pattern recognition learning.  Avtomat. i Telemekh. 25 (6), 937–945 (1964).
  7. El Moutaouakil K., El Ouissari A., Touhafi A., Aharrane N.  An Improved Density Based Support Vector Machine (DBSVM).  2020 5th International Conference on Cloud Computing and Artificial Intelligence: Technologies and Applications (CloudTech). 1–7 (2020).
  8. Mercer J.  XVI. Functions of positive and negative type, and their connection the theory of integral equations.  Philosophical Transactions of  The Royal Society of London. Series A. 209 (441–458), 415–446 (1909).
  9. Mangasarian O. L.  Generalized Support Vector Machines.  Advanced in Large Margin Classifiers. 135–146 (2000).
  10. Lin C. F.,  Wang S. D.  Fuzzy support vector machines.  IEEE transactions on neural networks. 13 (2), 464–471 (2002).
  11. Schölkopf B., Smola A. J., Williamson R. C., Bartlett P. L.  New support vector algorithms.  Neural computation. 12 (5), 1207–1245 (2000).
  12. Suykens J. A. K., Vandewalle J.  Least squares support vector machine classifiers.  Neural processing letters. 9 (3), 293-300 (1999).
  13. Schölkopf B., Platt J. C., Shawe-Taylor J., Smola A. J., Williamson R. C.  Estimating the support of a high-dimensional distribution.  Neural computation. 13 (7), 1443–1471 (2001).
  14. Bi J., Zhang T.  Support vector classification with input data uncertainty.  Advances in neural information processing systems. 161–168 (2005).
  15. Yang X., Song Q., Cao A.  Weighted support vector machine for data classification.  Proceedings. 2005 IEEE International Joint Conference on Neural Networks. 2, 859–864 (2005).
  16. Bi J., Vapnik V. N.  Learning with rigorous support vector machines.  Learning Theory and Kernel Machines. 243–257 (2003).
  17. Tang Y., Jin B., Sun Y., Zhang Y. Q.  Granular support vector machines for medical binary classification problems.  2004 Symposium on Computational Intelligence in Bioinformatics and Computational Biology. 73–78 (2004).
  18. Lee Y. J., Mangasarian O. L.  SSVM: A smooth support vector machine for classification. Computational optimization and Applications. 20 (1), 5–22 (2001).
  19. Lee Y. J., Mangasarian O. L.  RSVM: Reduced support vector machines.  Proceedings of the 2001 SIAM International Conference on Data Mining. 1–17 (2001).
  20. Schölkopf B., Smola A. J., Bach F.  Learning with kernels: support vector machines, regularization, optimization, and beyond. MIT press (2002).
  21. Mangasarian O. L., Wild E. W.  Proximal support vector machine classifiers.  Proceedings KDD-2001: Knowlborder discovery and data mining (2001).
  22. Mangasarian O. L., Wild E. W.  Multisurface proximal support vector machine classification via generalized eigenvalues.  IEEE transactions on pattern analysis and machine intelligence. 28 (1), 69–74 (2005).
  23. Khemchandani R., Chandra S.  Twin support vector machines for pattern classification.  IEEE Transactions on pattern analysis and machine intelligence. 29 (5), 905–910 (2007).
  24. Cortes, C.,  Vapnik, V. Support-vector networks. Machine learning, 20(3), 273-297 (1995).
  25. Wang Y., Wang S., Lai K. K.  A new fuzzy support vector machine to evaluate credit risk.  IEEE Transactions on Fuzzy Systems. 13 (6), 820–831 (2005).
  26. Huang H. P., Liu Y. H.  Fuzzy support vector machines for pattern recognition and data mining.  Int. J. Fuzzy Syst. 4, 826–835 (2002).
  27. Batuwita R., Palade V.  FSVM-CIL: fuzzy support vector machines for class imbalance learning. IEEE Transactions on Fuzzy Systems. 18 (3), 558–571 (2010).
  28. Yu H., Sun C., Yang X., Zheng S., Zou H.  Fuzzy support vector machine with relative density information for classifying imbalanced data.  IEEE transactions on fuzzy systems. 27 (12), 2353–2367 (2019).
  29. Khanam J. J., Foo S. Y.  A comparison of machine learning algorithms for diabetes prediction.  ICT Express. (2021).
  30. Tigga N. P., Garg S.  Prediction of type 2 diabetes using machine learning classification methods.  Procedia Computer Science. 167, 706–716 (2020).
  31. Shuja M., Mittal S., Zaman M.  Effective prediction of type ii diabetes mellitus using data mining classifiers and SMOTE.  Advances in computing and intelligent systems. 195–211 (2020).
  32. Devi R. D. H., Bai A., Nagarajan N.  A novel hybrid approach for diagnosing diabetes mellitus using farthest first and support vector machine algorithms.  Obesity Medicine. 17, 100152 (2020).