Early-Stage Detection of Alzheimer’s Disease Using Modified Firefly Based Ensemble Classification Model

Alzheimer's disease (AD) is a progressive neurodegenerative disorder characterized by cognitive decline affecting millions of people globally. Early detection is crucial to managing the disease effectively, enabling timely intervention to slow its progression. Most conventional machine learning approaches have been explored to enhance early prediction; however, these techniques often suffer from issues like overfitting, complex feature selection, and the inability to handle large datasets efficiently. To address the challenges of early AD prediction in the proposed model, the Hybrid Firefly optimized feature selection along with One Class SVM was used to achieve a better quality of Biomarker data and further integrate it with a customized ensemble classification model to improve the prediction rate. The proposed model performs better than conventional models and is measured in terms of accuracy, recall, precision and F1 score.