<![CDATA[Alzheimer’s disease is a progressive neurodegenerative disorder characterized by cognitive decline and impaired daily functioning, particularly among the elderly population. The increasing global prevalence of Alzheimer’s disease highlights the need for accurate, efficient, and accessible early detection methods. This study aims to analyze feature importance in predicting Alzheimer’s disease using the Random Forest algorithm. The dataset used is secondary data obtained from Kaggle, consisting of 2,148 patient records with 35 features covering demographic, medical, cognitive, and functional aspects. The research methodology includes data preprocessing, class imbalance handling using SMOTE, feature selection with SelectKBest, and model training and evaluation using Random Forest with K-Fold cross-validation. The results demonstrate that the Random Forest model achieved excellent performance with an accuracy of 94%, and balanced precision and recall values of 0.94. Feature importance analysis reveals that Functional Assessment, Activities of Daily Living (ADL), and Mini-Mental State Examination (MMSE) are the most influential predictors of Alzheimer’s disease. These findings indicate that cognitive and functional indicators play a more significant role in early Alzheimer’s detection than other medical factors. This study is expected to contribute to the development of effective and interpretable medical decision support systems for early Alzheimer’s disease detection.]]>
