<![CDATA[Environmental parameter prediction plays an essential role in supporting weather monitoring and data-driven decision-making, particularly in urban areas. However, prediction accuracy is often limited by a model’s ability to capture the interrelationships among environmental parameters. This study aims to analyze and compare the performance of two Long Short-Term Memory (LSTM) approaches Multivariate and Univariate in predicting air temperature as the dependent variable. In the Multivariate model, temperature prediction is influenced by other independent variables such as humidity, pressure, and altitude, whereas in the Univariate model, temperature prediction is based solely on its historical data. The model architecture consists of three main layers an input layer, two hidden layers, and an output layer. Model performance was evaluated using Root Mean Square Error (RMSE), Mean Square Error (MSE), and Mean Absolute Percentage Error (MAPE). The experimental results show that the multivariate LSTM model produces lower error values for temperature and pressure parameters, while the univariate LSTM model performs better for humidity and altitude. Therefore, the multivariate model is more suitable when the interrelationships among environmental parameters significantly influence prediction outcomes.]]>
Copyrights © 2025
