Alzheimer's Disease (AD) poses a significant global health challenge that necessitates early detection to find suitable treatments. Traditional detection methods face numerous obstacles, including the absence of definitive tests and early symptoms that may be misinterpreted as part of normal aging. While machine learning (ML) techniques are increasingly applied in detecting AD, few studies examine the impact of social, medical, and lifestyle factors on the onset of AD. This study underscores the importance of these features in improving diagnostic accuracy. An exploratory data analysis is conducted to explore the relationships between AD and these factors, followed by the application of random forest techniques to assess the significance of each factor and detect the presence of AD. The findings reveal that feature such as functional assessments, diet quality, and cognitive evaluations play a crucial role in AD development. Utilizing these insights, an ML-based detection model had been developed that achieved 95.55% accuracy, 96.50% precision, 90.19% recall, and a 93.08% F1 score, emphasizing the necessity of targeted feature selection to improve early detection and intervention strategies for AD management.