Researcher(s)
- Jared Bryant, Computer Science, University of Delaware
Faculty Mentor(s)
- Yixiang Deng, Computer and Information Science, University of Delaware
Abstract
Breast cancer is the most common cancer among women worldwide, with early detection crucial for improving survival. In this work, we train a Random Forest classifier on a Breast Cancer dataset to answer: How accurately can we distinguish malignant from benign tumors, and which measurements are most influential? Our tuned model achieved very high accuracy on the test set, missing only one malignant case. Using k‑fold cross‑validation and permutation‑importance testing, at this time I’ve identified tumor area, tumor texture, and concave‑points irregularity as the top predictors of malignancy.