 Our proposed approach combines machine learning algorithms and a long short-term memory, LSTM, based neural network to construct forecasting models for short to medium-term aggregate load forecasting. We use wrapper and embedded feature selection methods to choose the best features from the dataset and then optimize the LSTM models parameters using genetic algorithm, GA. This resulted in a decrease in mean absolute error, MAE, and root mean square error, RNSE, for medium to long-range forecasting for a wider metropolitan area. This article was authored by Sula Baoktif, Ali Faiz, Ali Auni, and others.