1Assistant Professor, Department of ECE, V.S.B. College of Engineering Technical Campus, Coimbatore, India. Email: narmadha.asn@gmail.com
2Associate Professor, Department of Biomedical Engineering, Erode Sengunthar Engineering College, Thudupathi, Erode, Tamilnadu. Email: sarithabme89@gmail.com
3UG Student, Department of ECE, V.S.B. College of Engineering Technical Campus, Coimbatore, India. Email: deepikasriravichandran@gmail.com
4UG Student, Department of ECE, V.S.B. College of Engineering Technical Campus, Coimbatore, India. Email: durgadeviramasamy2005@gmail.com
5UG Student, Department of ECE, V.S.B. College of Engineering Technical Campus, Coimbatore, India. Email: harinim4117@gmail.com
6UG Student, Department of ECE, V.S.B. College of Engineering Technical Campus, Coimbatore, India. Email: charulatha4747@gmail.com
Breast cancer is one of the leading causes of death among women worldwide, and early diagnosis plays a critical role in increasing patient survival rates. Currently, the traditional diagnostic methods are time-consuming and inefficient, depending either on manual interpretation of clinical or diagnostic information, or other factors related to judgment -- such as how cells look under a microscope and characteristic changes in human tissue. In recent years, machine learning techniques have emerged as effective tools for automated disease diagnosis because they can recognize the many patterns and nuances of medical datasets. This work uses machine learning methods to classify broad category of breast cancer tumours with Wisconsin Breast Cancer Datasets using a comprehensive framework. We use multiple supervised learning techniques implemented in Python, Ipython notebooks, Pandas packages etc.: Logistic Regression, Random Forest, Gradient Boosting, XGBoost, CatBoost. In order to further improve prediction accuracy and strengthen resilience, we proposed a stacking ensemble scheme in which the predictions made by individual learners are used as input to a meta-learner. Experimental results show that the combined model performs better in terms of accuracy, precision, recall and F1-score than any of its component classifiers. By so doing, it also avoids over-fitting and leads to greater generalization. In addition, the proposed system illustrates how machine learning with a combination of ensembles can serve as intelligent clinical decision support tools to help guide human professionals in their work. It also contributes to timely diagnosis of early breast cancer and efficient analysis.
Keywords: Breast Cancer Classification, Machine Learning, Ensemble Learning, Medical Diagnosis, Predictive Analytics.
How to cite this article: Narmadha AS, Saritha B, Deepika sri R, Durgadevi R, Harini M, Charulatha SE. Breast Cancer Classification Using Machine Learning: A Comprehensive Analysis of Predictive Models. Int J Drug Deliv Technol. 2026;16(19s): 8-16. DOI: 10.25258/ijddt.16.19s.2
Source of support: Nil.
Conflict of interest: None