Big Data Analytics for Climate Change Prediction using Hadoop and ML Models

Climate change poses a significant global challenge, necessitating accurate prediction models to assess its impact and formulate mitigation strategies. This research focuses on developing an efficient big data framework using Hadoop and machine learning models to analyze and predict climate trends. By leveraging large-scale climate datasets, the study implements Prophet, Linear Regression, and Random Forest algorithms to forecast key climate parameters such as temperature, humidity, and CO₂ levels. A comparative analysis of these models reveals that Linear Regression and Random Forest demonstrate the highest accuracy with an R² score of 0.97, making them effective tools for climate prediction. Additionally, a Streamlit-based AI Climate Prediction Dashboard is developed to provide real-time weather insights, historical analysis, and future climate projections. The proposed framework offers actionable insights for policymakers, researchers, and industries to implement informed climate adaptation strategies. Future work will focus on integrating deep learning models, real-time data processing, and hybrid AI techniques to further improve prediction accuracy and climate impact assessment.