Scholarlink

This study compares the performance of Kolmogorov-Arnold Networks (KAN) with traditional machine learning models in predicting breast cancer survival. Using the SEER Breast Cancer Dataset, we evaluate KAN against decision trees, random forests, logistic regression, and gradient boosting machines. Our analysis focuses on predictive accuracy and the impact of data augmentation techniques. Results demonstrate that KAN offers competitive performance compared to traditional models, with unique strengths in handling imbalanced datasets. This research contributes to the ongoing exploration of novel machine learning architectures in healthcare applications.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Comparative Analysis of Kolmogorov-Arnold Networks and Traditional Machine Learning Models for Breast Cancer Prognosis

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Haseebullah Jumakhan,
  • Amir Mirzaeinia

摘要

This study compares the performance of Kolmogorov-Arnold Networks (KAN) with traditional machine learning models in predicting breast cancer survival. Using the SEER Breast Cancer Dataset, we evaluate KAN against decision trees, random forests, logistic regression, and gradient boosting machines. Our analysis focuses on predictive accuracy and the impact of data augmentation techniques. Results demonstrate that KAN offers competitive performance compared to traditional models, with unique strengths in handling imbalanced datasets. This research contributes to the ongoing exploration of novel machine learning architectures in healthcare applications.