Evaluating Machine Learning Algorithms for Early Detection of Alzheimer's Disease

Alzheimer's disease is a neurological condition that worsens over time and impairs cognitive function; therefore, early diagnosis is crucial to improve patient treatment. In this study, the Alzheimer's Prediction Dataset, comprising 74,283 records with 25 features such as demographic, genetic and lifestyle factors, is employed to compare Machine Learning (ML) techniques for predicting Alzheimer's disease. After the dataset is divided and features scaled, random forest, gradient boosting and logistic regression are trained. With a 73.93% prediction accuracy, Gradient Boosting has a better performance than random forest (72.45%) and logistic regression (71.66%). Accuracy, recall, F1-score and precision are the measures for assessing the model, which testifies to ability of Gradient Boosting to recognize complex patterns. The research indicates the ways in which ML can assist in diagnosis of Alzheimer's and highlights age as a key determinant. For greater accuracy, subsequent studies will investigate deep learning techniques and feature selection optimization.