magazinelogo

Journal of Applied Mathematics and Computation

ISSN Print: 2576-0645 Downloads: 156586 Total View: 1858221
Frequency: quarterly ISSN Online: 2576-0653 CODEN: JAMCEZ
Email: jamc@hillpublisher.com
Article Open Access http://dx.doi.org/10.26855/jamc.2020.06.002

Application of an Improved FSVM Algorithm in Breast Cancer Diagnosis

Jinzhi Zhou *, Jing Huang, Linwen Zheng

School of Information Engineering, Southwest University of Science and Technology, Mianyang, Sichuan, China.

*Corresponding author: Jinzhi Zhou, School of Information Engineering, Southwest University of Science and Technology, Mianyang, Sichuan, China.

Published: May 15,2020

Abstract

The traditional method of constructing fuzzy support vector machine (FSVM) membership function based on the Euclidean distance between the sample and the class center treats all features equally, without considering the effect of different features on the distance between the sample and the class center. To solve this problem, a fuzzy support vector machine classification algorithm based on Relief-F feature weighting is proposed. First, the weights of each feature are calculated by the Relief-F algorithm and the features with smaller weights are removed; then the feature weights are used to calculate the weighted Euclidean distance from the sample to the center of the class; finally, the membership function is constructed based on the weighted Euclidean distance. This method considers the influence of feature importance on the classification effect, and removes the features with less weight through the weight threshold, thereby reducing the dimension of the data and improving the classification accuracy and training efficiency. And apply in the diagnosis of breast cancer. After theoretical analysis and experimental data verification, this method improves the performance of classification prediction compared to traditional support vector machines (SVM). The results are also competitive with the latest methods and have advantages in medical diagnostic applications.

References

[1] Vapnik V. (1995). The Nature of Statistical Learning Theory. New York: SpringerVerlag, 1995: 91-188.

[2] C. Burges. (1998). A tutorial on support vector machines for pattern recognition [J], Data Mining and Knowledge Discovery, vol. 2, no. 2, pp. 121-167.

[3] Zhang Xuegong. (2006). On Statistical Learning Theory and Support Vector Machines [J]. Acta Automatica Sinica, 2000, 26 (1): 32-41.

[4] Alexandre R. Mello, Marcelo R. Stemmer, Alessandro L. Koerich. (2020). Incremental and decremental fuzzy bounded twin support vector machine[J]. Elsevier Inc., p. 526.

[5] Tahereh Bahraini,Saeedeh Ghazi,Hadi Sadoghi Yazdi. (2020). Toward optimum fuzzy support vector machines using error distribution[J]. Elsevier Ltd, p. 90.

[6] LIN Chun-fu, WANG Sheng-de. (2002). Fuzzy support vector machines [J]. IEEE Transactions on Neural Networks, 3(2): 464-471.

[7] Zhang Xiang, Xiao Xiao-ling, and Xu Guang-you. (2006). Fuzzy support vector machines based on affinity among samples. Journal of Software, 17(5): 951-958.

[8] Fan. X, Du. S, Wu. T. (2003). “Weighted Support Vector Machine Based Classification Algorithm for Uneven Class Size Problems”. Chinese Journal of Image and Graphics. Vol 8, No.9, pp. 1037-1042.

[9] Li Xiaohui, Zhu Li, Wu Jian. (2017). Target recognition of millimeter wave radiometer based on fuzzy support vector machine [J]. Chinese Journal of Microwaves, 33 (S1): 203-206.

[10] Liu Xingchang. (2019). Research on Xixia Character Recognition Based on Improved Fuzzy Support Vector Machine [D]. Lanzhou Jiaotong University.

[11] Zhao Hui and Rong Li-li. (2005). Classification algorithm based on combined support vector machines and its application in text categorization. Mini-Micro Systems, 26(10): 1816-1820.

[12] Tseng Yi-Ju, Huang Chuan-En, Wen Chiao-Ni, et al. (2019). Predicting breast cancer metastasis by using serum biomarkers and clinicopathological data with machine learning technologies. [J]. International journal of medical informatics.

[13] Kho Esther, de Boer Lisanne L, Van de Vijver Koen, et al. (2019). Hyperspectral imaging for resection margin assessment during cancer surgery. [J]. Clinical Cancer Research : An Official Journal of the American Association for Cancer Research.

[14] Kira K, Rendell L. (1992). A practical approach to feature selection [C]//Proceedings of the 9th International Workshop on Machine Leaning. San Francisco: Mogran Kaufmann,1992: 249-256.

[15] Kononenko I. (1994). Estimating at tributes: Analysis and extensions of RELIEF [J]. Lecture Notes in Computer Seience. 784:171-182.

How to cite this paper

Application of an Improved FSVM Algorithm in Breast Cancer Diagnosis

How to cite this paper: Jinzhi Zhou, Jing Huang, Linwen Zheng. (2020) Application of an Improved FSVM Algorithm in Breast Cancer Diagnosis. Journal of Applied Mathematics and Computation, 4(2), 18-25.

DOI: http://dx.doi.org/10.26855/jamc.2020.06.002