Enhancing Alzheimer's Disease Detection Using Optimized Attribute Selection and Random Forest Classifier for Improved Accuracy

This research aims at improving the identification of Alzheimer's dementia by using the optimized attribute selection method and preliminary diagnosis data-driven learning algorithm, namely Random Forest Classifier. Alzheimer's is a progressive neurological disorder exhibiting barriers in preliminary diagnosis primarily due to the high density and intricate nature of medical information. In this study, attribute selection techniques are used to minimize the data dimensionality, by selecting only key features for classification with low computational complexity and improved model accuracy. The Random Forest Classifier is used because it is accurate, can process big data and because of its diverse uses. Through selecting the most relevant features of the model, it provides a prominent level of precise and acute identification of the early stage of Alzheimer's and could significantly enhance the medical treatment effects. The outcomes from this study indicate that feature optimization can be used interchangeably with preliminary diagnosis data-driven learning algorithms for accurate disease prediction.