IJRCIT- International Journal of Research in Computer & Information Technology

Archives of International Journal of Research in Computer & Information Technology(IJRCIT)

Volume 10 Issue 3 June 2025

1. Ensemble-Driven Machine Learning Models for Robust Breast Cancer Recurrence Prediction

AUTHOR NAME : Prachi Damodhar Shahare, Dr. Abha Mahalwar

ABSTRACT : Breast cancer recurrence remains a major clinical challenge despite advances in early diagnosis and treatment. Accurate prediction of recurrence can support personalized follow-up strategies and improve patient outcomes. This study investigates the effectiveness of multiple machine learning models for breast cancer recurrence prediction using clinical and tumor-related features. Decision Tree, Logistic Regression, Support Vector Machine (RBF), Random Forest, and a soft-voting Ensemble model are developed and evaluated. Model performance is assessed using several evaluation parameters. Experimental results demonstrate that while Logistic Regression achieves the highest ROC–AUC (0.702), the proposed Ensemble model provides the most balanced performance, yielding superior recall (0.6118), F1-score (0.5828), and MCC (0.6257). Calibration analysis further confirms improved probability reliability for the ensemble approach. The findings highlight the benefit of ensemble learning for robust and clinically reliable breast cancer recurrence prediction.

Download