Comprehensive examination and comparative analysis of machine learning algorithms for breast cancer prediction
Ajay Kumar, Debolina Ghosh, Dashrath Mahto, Jay Prakash Singh · 2025
Breast cancer stands as the predominant form of cancer among females globally, encompassing India, where the heightened prevalence of advanced stages at diagnosis, alongside escalating incidence, and mortality rates, underscores the critical importance of comprehending cancer literacy among women. Besides an early breast cancer detection and risk assessment through a comprehensive comparative analysis of machine learning algorithms employing the Wisconsin Breast Cancer dataset. The investigation meticulously evaluates the efficacy of these algorithms in terms of accuracy, interpretability, and computational efficiency. By pinpointing standout models, this study accentuates their clinical relevance, thereby highlighting the capacity of machine learning to empower healthcare practitioners in rendering well-informed decisions and elevating patient prognosis. The notable accuracy rates validate the efficacy of ML models in this critical domain. K-Neighbors (k-NN) achieved 96.48%, Ada Boost(ABC) 95.47%, and Decision Tree (DT) 96.98% accuracy, emphasizing model diversity and healthcare applicability. Support Vector Classifier (SVC) 91.95% and Gradient Boosting (GBC) 97.48% contribute further insights. Despite slightly lower accuracy, SVC demonstrates model versatility. Ultimately, Logistic Regression (LRC) emerges as the most efficient approach for classifying tumors, achieving 98.49% accuracy, surpassing all other ML algorithms. These findings underscore the paradigm-shifting role of machine learning in medical diagnostics, elucidating its pivotal contributions to the continual combat against breast cancer.