In data processing applications going from picture recognizable proof to time series expectation, Artificial Neural Networks (ANNs) have made astonishing progress. The accessibility of colossal datasets for preparing, as well as the rising intricacy of the models, might be added to the achievement. Unfortunately, just a set number of models are accessible for preparing in certain applications. With high-intricacy models, less preparation tests increment the gamble of over-fitting and unfortunate speculation. Likewise, complex models with an enormous number of teachable boundaries need more energy to prepare and improve than easier models. This study clearly demonstrates the importance of ANNs for writing gathering in the early stages of Alzheimer's disease (ES-AD).