Stacking Based Ensemble Learning Framework for Lung Cancer Prediction

Abstract

This study proposes a novel stacking-based ensemble framework for predicting lung cancer, leveraging the power of machine learning algorithms. The proposed framework utilizes multiple base models that are trained on different subsets of the data, along with a meta-learner that combines the outputs of the base models to generate the final prediction. The base models include Support Vector Machines, k-Nearest Neighbors, Extra Trees Classifier, Random Forests, Gradient Boosting Machines, etc, while the meta-learner is a Decision Tree model. The proposed framework was evaluated using a publicly available lung cancer dataset, containing clinical information of patients along with their cancer diagnosis. The results show that the stacking-based ensemble framework outperforms each of the individual base models, achieving an accuracy of 95%. Furthermore, the study conducted an extensive feature selection analysis to identify the most informative features for lung cancer prediction.