This document summarizes a research paper that proposes a hybrid model called ISCOA-LSTM for building energy consumption forecasting. ISCOA-LSTM combines long short-term memory (LSTM) networks with an improved sine cosine optimization algorithm (ISCOA) to optimize the LSTM hyperparameters. Experimental results on energy consumption data from an academic building show that ISCOA-LSTM achieves more accurate short, mid, and long-term energy forecasts compared to other models, as measured by lower errors.