An Initial Survey on CO2 Emission Analysis and Prediction using Algorithms

The progressive increase in the emissions of carbon dioxide (CO2), one of the commonly recognized greenhouse gases, has been recognized as a prominent contributing factor to global warming and climate change. This research utilizes a systematic method in assessment and modelling of CO2 emissions by extracting a publicly available dataset from the Kaggle database. This examines emissions data at the country level, using data mining techniques to detect unusual observations, development over time, and regularities. Through rigorous exploratory data analysis (EDA) and subsequent analysis, very useful information is gained regarding the spatial patterns of CO2 emissions and their correlates. Class imbalance is addressed by applying the Synthetic Minority Over- sampling Technique (SMOTE) to the dataset. The machine learning models developed in this study will support decision-makers in formulating effective strategies to reduce emissions, aligning with the specific objectives set by policymakers. Such initiatives greatly bring about protection of the environment and facilitate in attaining sustainability goals in the world.