Mohit Pant, Rashmi Pant

DOI: https://doi.org/10.5281/zenodo.22308358

Breast cancer is a major health concern, and early diagnosis is crucial for treatment. In this paper we present a machine learning-based approach for classifying breast cancer using the Wisconsin Diagnostic Breast Cancer (WDBC) dataset. CfsSubsetEval with GreedyStepwise is used to select the most relevant features and reduce redundancy. The selected features are evaluated using Logistic Regression, Multilayer Perceptron (MLP) classifiers and Sequential Minimal Optimization (SMO). The models are compared using accuracy, precision, sensitivity, specificity, F1-score, and ROC-AUC. The goal of this study is to identify an efficient classifier and a compact feature subset for reliable breast cancer classification.

Breast cancer, Multilayer Perceptron, Regression, Sequential Minimal Optimization

This paper has investigated the use of machine learning for breast cancer classification using the Breast Cancer Wisconsin Diagnostic (WDBC) dataset. Logistic regression, multilayer perceptron (MLP), and SMO were applied before and after feature selection with CfsSubsetEval with GreedyStepwise. The results show that feature selection can reduce redundant and less informative attributes and still deliver good classification performance. Before the feature selection, SMO had the highest accuracy of 98.25% and MLP and Logistic Regression had 95.91% and 90.64%, respectively. Logistic Regression was improved with CfsSubsetEval and GreedyStepwise and achieved 98.24% accuracy, 98.60% precision, 96.60% sensitivity, 99.20% specificity, 97.60% F1-score, and 0.998 ROC-AUC. MLP and SMO had very good classification performance and had 97.49% and 97.24% accuracy, respectively. The correlation-based feature selection combined with GreedyStepwise search can identify the relevant and non-redundant features for breast cancer diagnosis. The better performance of Logistic Regression shows that a simple and interpretable classifier can accomplish competitive results if the right feature subset is used. Therefore, our proposed CfsSubsetEval–GreedyStepwise feature selection framework combined with Logistic Regression can be considered a promising approach for the development of an efficient breast cancer classification system.

[1] W. H. Wolberg, W. N. Street, and O. L. Mangasarian, “Breast Cancer Wisconsin (Diagnostic),” UCI Machine Learning Repository, 1995.
[2] W. H. Wolberg, W. N. Street, and O. L. Mangasarian, “Machine learning techniques to diagnose breast cancer from image-processed nuclear features of fine needle aspirates,” Cancer Letters, vol. 77, no. 2–3, pp. 163–171, 1994.
[3] I. Ozcan, H. Aydin, and A. Cetinkaya, “Comparison of classification success rates of different machine learning algorithms in the diagnosis of breast cancer,” Asian Pacific Journal of Cancer Prevention, vol. 23, no. 10, pp. 3287–3297, 2022.
[4] A. F. M. Agarap, “On breast cancer detection: An application of machine learning algorithms on the Wisconsin diagnostic dataset,” in Proc. 2nd Int. Conf. Machine Learning and Soft Computing, 2018, pp. 5–9.
[5] M.-L. Huang, Y.-H. Hung, and W.-Y. Chen, “Neural network classifier with entropy based feature selection on breast cancer diagnosis,” Journal of Medical Systems, vol. 34, no. 5, pp. 865–873, 2010.
[6] Mert, A., Kılıç, N., Bilgili, E., & Akan, A. (2015). Breast cancer detection with reduced feature set. Computational and mathematical methods in medicine, 2015(1), 265138.
[7] Ibrahim, S., Nazir, S., & Velastin, S. A. (2021). Feature selection using correlation analysis and principal component analysis for accurate breast cancer diagnosis. Journal of imaging, 7(11), 225.
[8] I. H. Witten, E. Frank, M. A. Hall, and C. J. Pal, Data Mining: Practical Machine Learning Tools and Techniques, 4th ed. Morgan Kaufmann, 2016.
[9] Karatza, P., Dalakleidi, K., Athanasiou, M., & Nikita, K. S. (2021, November). Interpretability methods of machine learning algorithms with applications in breast cancer diagnosis. In 2021 43rd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC) (pp. 2310-2313). IEEE.

Mohit Pant, Rashmi Pant (2026). Comparison of Machine Learning Models for Detecting Breast Cancer Using WEKA Software. International Journal of Computer Science Engineering Techniques, 10(5), 139–144. ISSN: 2455-135X. DOI: https://doi.org/10.5281/zenodo.22308358