Alzheimer’s disease is a unique neurologic condition that causes the mind to decay and synapses to graze on the leftovers. Around 45 million people are affected by this disease. Many investigations on the automatic diagnosis of Alzheimer’s Disease (AD) utilizing various methodologies have been carried out in recent years. However, detecting symptoms as early as possible (predetection) is critical, because disease-modifying medications are most successful when given early in the course of the disease before permanent brain damage occurs. As a result, using automated approaches to predict AD symptoms from such data is critical. Using the dataset from the Open Access Series of Imaging Studies (OASIS) the system was built. The data was examined and used in various machine learning models. For prediction, decision trees, support vector machines, random forests and logistic regression were applied. Except for a manually controlled neural network, which takes significantly longer to train, the SVM surpasses all of them with an accuracy of 88.88%.