Boosting the Accuracy of Weather Forecasting Through a Machine Learning Approach: Case Study of CNN, GRU, RNN, and Random Forest

Abstract

The accuracy in predicting the weather is important for agriculture, disaster management, and energy planning sectors. Typical weather forecasting techniques are not very successful with complicated atmospheric systems, which has led to greater interest in machine learning solutions. In this research, machine learning techniques including Random Forest, Gated Recurrent Unit (GRU), Recurrent Neural Network (RNN) and Convolutional Neural Network (CNN) are utilized in their respective domains as temperature trend forecasting models, using historical weather data. The results emphasize the effectiveness of deep learning techniques, especially CNN, for capturing time series data with complex interdependencies such as weather data over time.