Fuzzy PCA and Support Vector Machines for Breast Cancer Classification

Abstract

Breast cancer is the leading cause of death among women in the world and early detection can increase the chance of survival for the patients. However, expert system and machine learning diagnosis are burdened with the presence of irrelevant data and noise which can reduce the accuracy of prediction and increase computational time. In this paper, Fuzzy Principle Component Analysis (FPCA) and Support Vector Machines (SVM) are proposed for the classification of breast cancer dataset. Experimental results on public breast cancer dataset show that the proposed method FPCA-SVM outperformed the benchmark models in terms of accuracy, specificity, and sensitivity and AUC value. The proposed model can assist doctors and medical practitioners for an early detection of breast cancer.

Principal component analysis based on intuitionistic fuzzy random variables

“…Parsaie et al (2018) developed an adaptive neuro-fuzzy inference system model to predict the longitudinal dispersion coefficient based on the PCA results. Dzulkalnine et al (2018) employed the fuzzy PCA and support vector machines for classification of breast cancer data. For more on recent applications of fuzzy PCA, see also (Abhishek et al 2017;Tao et al 2019).…”