Now a days even though there are different methods to diagnose Alzheimers disease ,it is always been challenging to the doctors for proper diagnosis and planning of appropriate treatment of this disease. The manual spotting of Alzheimers disease from MRI images is driven by many factors and may differ from experts depending on their expertise in the diagnosis of the disease. The automatic and accurate identification of Alzheimers disease from MRI images helps to eliminate the above issues and deliver better results. Alzheimers disease is an irrevocable brain disease that steadily destroys brain cells and hence resulting in permanent memory losses. To overcome this situation, a new algorithm is developed by combining different point detection and feature extraction methods for early prediction of various stages of Alzheimer's disease. This new method is integrated with different classifiers like k Nearest Neighbor and Random Forest classifiers. The classification accuracy and performance are evaluated and analyzed for both classifiers. The new proposed method provides an accuracy rate of 98.3% when k Nearest Neighbor classifier is used and gives an accuracy rate of 97.12% when Random Forest classifier is used. From the results the new proposed algorithm is found to be more efficient than the existing algorithms since it uses a combination of multiple feature extraction methods when compared to the methods which uses single feature extraction method.