From Data to Diagnosis: Harnessing Machine Learning for Early Breast Cancer Detection

One of the most prevalent and deadly illnesses affecting women worldwide is breast cancer. Improving patient outcomes and survival rates requires a prompt diagnosis and early detection of breast cancer. In this work, we use the Breast Cancer Wisconsin (Diagnostic) Data Set and the Breast Cancer Wisconsin - benign or malignant dataset to propose an effective machine learning framework for the early detection of breast cancer. Employing a number of classification models on the dataset to be assessed in terms of accuracy and mean absolute error (MAE) include Random Forest (RF), Decision Tree (DT), AdaBoost (AD), Naive Bayes (NB), and Support Vector Machine (SVM). Breast cancer prediction is made possible by these models, which are trained on both datasets and identify patterns in breast cancer detection. Naive Bayes unexpectedly achieved the highest test accuracy at 97.37%, despite having weaker other metrics. Random Forest and SVM produced the best results on the first breast cancer dataset, both with 96.49% accuracy. Random Forest won the second dataset with 97.78% accuracy, followed by SVM and Naive Bayes at 95.56%. This demonstrates how machine learning can be a useful tool for early breast cancer detection. More research and backing from medical professionals could make these models even more accurate and useful for diagnosing problems in the real world.