Tree-Based and Machine Learning Algorithm Analysis for Breast Cancer Classification

Area of Science:

• Oncology
• Medical Informatics
• Machine Learning

Background:

• Breast cancer (BC) is a significant global health concern, being the second leading cause of death.
• BC is characterized by genetic mutations and physical changes like pain, size fluctuations, and altered skin texture.
• Accurate diagnosis is crucial for effective treatment and improved patient outcomes.

Purpose of the Study:

• To evaluate and compare the performance of four machine learning algorithms for breast cancer classification.
• To identify the most accurate classifier for distinguishing between benign and malignant breast tumors using the Wisconsin Diagnostic Breast Cancer (WDBC) dataset.

Main Methods:

• The study utilized the Wisconsin Diagnostic Breast Cancer (WDBC) dataset, sourced from fine-needle aspiration biopsies.
• Four machine learning algorithms were implemented: Multilayer Perceptron (MLP), K-Nearest Neighbor (KNN), Genetic Programming (GP), and Random Forest (RF).
• Performance was evaluated based on classification accuracy.

Main Results:

• Random Forest (RF) demonstrated the highest classification accuracy at 96.24%.
• RF significantly outperformed MLP, KNN, and GP in classifying breast cancer cases.
• The results highlight RF's effectiveness in diagnosing breast cancer from biopsy data.

Conclusions:

• Random Forest is a highly effective machine learning model for accurate breast cancer classification.
• The study validates the utility of machine learning in improving diagnostic accuracy for breast cancer.
• Further research can explore ensemble methods or deep learning for even greater precision.