Robust Stacked Ensemble Model for Lung Cancer Diagnosis

Akbas A., Buyrukoglu G., Buyrukoglu S.

9th International Conference on Computer Science and Engineering, UBMK 2024, Antalya, Turkey, 26 - 28 October 2024, pp.436-440, (Full Text) identifier

  • Publication Type: Conference Paper / Full Text
  • Doi Number: 10.1109/ubmk63289.2024.10773466
  • City: Antalya
  • Country: Turkey
  • Page Numbers: pp.436-440
  • Keywords: Chi-Square, Feature Selection Techniques, Lung Cancer Diagnosis, Machine Learning, ReliefF, Stacked Ensemble Model
  • Abdullah Gül University Affiliated: No

Abstract

In this paper, we introduce a stacked ensemble model for lung cancer diagnosis, combining Random Forest, Support Vector Machine, and Artificial Neural Networks. Our model, validated on two datasets and achieving diagnostic accuracies of 92.6% and 88.9%, outperforms traditional algorithms. We also highlight the model's robustness through advanced feature selection methods like ReliefF and Chi-Square. The results we achieved showcase the potential of machine learning techniques in medical diagnostics and contribute significantly to the field of computational oncology, offering a promising approach for early lung cancer detection and prognosis.