The project compared Machine Learning and Deep Learning Forecasting frameworks to determine which one provided more accurate and optimum models in order to estimate sales based on historical data. Four (4) Machine Learning models and three (3) Deep Learning models were used for this purpose.

The objectives of this research were to apply these state of art algorithms to a process that has been performed in a very traditional way; learn and apply new Machine and Deep learning models that can be very effective in the organization; gain a better understanding of marketing campaigns and retail business as well as of the processes used to calculate the metrics involved in the process.