Breast Cancer Prediction using SVM with PCA Feature Selection Method

Abstract

Cancer has been characterized as one of the leading diseases that cause death in humans. Breast cancer, being a subtype of cancer, causes death in one out of every eight women worldwide. The solution to counter this is by conducting early and accurate diagnosis for faster treatment. To achieve such accuracy in a short span of time proves difficult with existing techniques. Also, the medical tests conducted in hospitals for detecting cancer is expensive and is difficult for any common man to afford. To counter these problems, in this paper, we use the concept of applying Support Vector machine a Machine Learning algorithm to predict whether a person is prone to breast cancer. We evaluate the performance of this algorithm by calculating its accuracy and apply a min-max scaling method so as to counter and overcome the problem of overfitting and outliers. After scaling of the dataset, we apply a feature selection method called Principle component analysis to improve the algorithms accuracy by decreasing the number of parameters. The final algorithm has improved accuracy with the absence of overfitting and outliers, thus this algorithm can be used to develop and build systems that can be deployed in clinics, hospitals and medical centers for early and quick diagnosis of breast cancer. The training dataset is from the University of Wisconsin (UCI) Machine Learning Repository which is used to evaluate the performance of the Support vector machine by calculating its accuracy.

A data-driven principal component analysis-support vector machine approach for breast cancer diagnosis: Comparison and application

“…The results for Yang and Xu (2019) and Yin et al (2016) are also slightly better than our results, because they do not use cross-validation and simply group some datasets into testsets, potentially making test dataset easier to distinguish. Yadav et al (2019) also used SVM classifier whose result is different from ours, it is also because of the partition of test dateset, they used 40% data as test dataset. We used 10-fold cross-validation, and the results are less random and more reliable.…”

“…With the comparison of the results between previous studies (Fan et al, 2011; Krishnan et al, 2010; Mert et al, 2014; Sweilam et al, 2010; Yadav et al, 2019; Yang et al, 2019) and this study in Table 7, it can be found that the 10-CV, PCA-SVM (RBF) used in this study has the highest accuracy (97.19%) except the best result (98.90%) of CBFDT, which was proposed by Fan et al (2011). They used all the attributes, while the results obtained in this paper is based on PCA-SVM using six principal components only.…”

Controlling and enhancing performance using QM-Window in queuing models

“…(Yadav & Sohani, 2019) [7] , utilized the MM1 queue model to estimate the performance of the food chain's services, and described how as the number of servers grows, so does the system's service quality. The "Queuing model as a Method of Queue resolution in Nigeria Banking Sector" is the subject of research by Anichebe (2019) [8] .…”

Simulación de Sistemas de Colas para Optimizar el Servicio al Cliente en el Sector de Restaurantes.

“…Esto perjudica a los clientes y frena la expansión de las empresas. Una empresa debe trabajar siempre para mejorar sus servicios, concentrarse en satisfacer las necesidades de los consumidores de la manera más eficaz posible y emplear una estrategia que funcione para la empresa si quiere tener éxito en el despiadado mercado actual (Yadav & Sohani, 2019).…”