A Comparative Study of Machine Learning Algorithms for Early Prediction of Alzheimer's Disease

Alzheimer's disease (AD) is one of the most common neurodegenerative disorders, leading to progressive impairment of cognitive function. There is an immediate need to diagnose AD early to intervene in a timely fashion that could slow up disease progression. Traditional diagnosis methods are often expensive and take a very long time, motivating the pursuit of ML as an alternative diagnosis approach. In terms of algorithms, the effectiveness of five types was compared: Random Forest, Gradient Boosting, Logistic Regression, Support Vector Machine (SVM), and k-Nearest Neighbors (k-NN) in predicting Alzheimer's disease. Different metrics such as accuracy, precision, recall, and F1-score were utilized using a patient dataset. Results show that Gradient Boosting achieved the highest accuracy at 95.81%, followed by Random Forest at 92.79%. The ensemble methods were seen to perform pretty well since they had the ability to manage complex patterns and noisy data. On the contrary, the performance was relatively less precise for k-NN (72.55%) due to the very high dimensionality of the dataset itself. The findings emphasize the requirement of the suitable machine learning approach that would be used for better predictive accuracy, particularly the earlier detection of AD.