A Novel Interpretable Lightweight Ensemble Learning Method for Static and Dynamic Medical and Healthcare Data Classification

In the medical field, efficient and accurate classification of sequential and structured data is crucially important and useful for early diagnosis and treatment. Traditional machine learning models struggle with the complexity and nonlinearity of dynamic datasets, whereas deep learning models, despite their effectiveness, require extensive resources and lack transparency. This paper proposes a novel lightweight ensemble framework integrating a parameterized SoftMax function with a non-parametric Random Forest method through a soft voting mechanism, supported by the Nonlinear AutoRegressive eXogenous (NARX) model and optimized using a forward orthogonal search and selection (FOSS) algorithm for feature selection. This innovative approach enhances the accuracy and robustness of classifiers for both static and dynamic medical datasets, while improving interpretability and computational efficiency. Extensive validation on various medical datasets demonstrates the model's superior performance and adaptability, offering a reliable solution for complex medical data scenarios. It is expected that the results achieved in this study paves the way for future innovations in medical data analysis and the broader application of artificial intelligence in healthcare.