Comparison of Breast Cancer Prediction Using Machine Learning

Abstract

Recent times, breast cancer has seen a concerning rise, affecting a significant proportion of women. To tackle this pressing issue, extensive research efforts have been dedicated to devising effective methodologies for early detection and prediction. Our proposed approach leverages techniques to predict potential risks also promote recent alert of breast cancer. What sets our approach apart is its practical applicability in real-world scenarios, offering a straightforward method for breast cancer prediction. We harnessed the power of four datasets hosted on the Kaggle platform and integrated various classifiers, including Decision Tree (DT), Random Forest (RF), Logistic Regression (LR), K-Nearest Classifier (KNN), among others, into our model. The results were promising, with the KNN achieving a noteworthy test accuracy of 81.14% for Dataset A, KNN of 97.2% for dataset B, KNN of 98.85% for dataset C and LR of 96.125% for dataset D. Furthermore, Bagging KNN also demonstrated accuracy matching this high standard of 99.42%. To further enhance performance, we implemented a range, including Bagging, Boosting, Stacking and Voting algorithms, optimizing each classifier with the best parameters through hyperparameter tuning. Through our experimental investigation, we not only contributed to the body of knowledge on breast cancer detection and prediction but also identified the KNNB (K-Nearest Classifier with Bagging) model as the most accurate, achieving an outstanding accuracy rate of 99.42% for breast cancer predictions. This research endeavors to provide invaluable insights into breast cancer management, offe