A Hybrid AI Framework for Climate Change Prediction and Mitigation Using Deep Learning and Reinforcement Learning
Authors:-Yashraj Misal, Omkar Ovhal, Aqdas Mirza, Mayuresh More, Mrs. Anuja S. Phapale
Abstract-This paper proposes a hybrid artificial intelligence (AI) modular system with reinforcement learning and deep learning to update climate change forecasting and mitigation planning. The system learns from multimodal data in satellite imagery, IoT sensors, and historical data to make accurate, real- time forecasts of key climate variables. Convolutional neural networks (CNN) and long-short-term memory (LSTM) models are employed to learn spatial and temporal patterns, and a Re- inforcement Learning (RL) module provides adaptive mitigation recommendations. Prediction accuracy and emission reduction are enhanced compared to traditional models, as shown in our experimental results. This paper contributes significantly to scalable and intelligent solutions to climate change issues.
