Our research work focuses on the development of framework of regional and national greenhouse gas emissions (GHG) inventory accounting and forecasting to support the effective mitigation strategies. By integrating data-driven methods, our work aims to understand GHG emissions patterns and inform policymaking at both local and national level. Our first project targets at developing comprehensive emissions inventories for three regions in Connecticut across six key sectors: mobile combustion, electricity consumption, solid waste, stationary combustion, agriculture, and wastewater treatment. This study uses Autoregressive Integrated Moving Average (ARIMA) model, an advanced statistical model, to forecast medium-term emissions trends and analyze the potential emissions shifts. In addition, a data-driven scenario discovery approach based on Patient Rule Induction Method (PRIM) was applied to search the most effective policy and strategy combinations for targeted emissions reduction across various sectors. Furthermore, our work focuses on the national level emissions mitigation efforts and unsupervised and supervised machine learning techniques were applied to analyze mobile GHG emissions pattern and influencing factors. The findings from those research offer value insights and actionable frameworks to inform current and future emissions targets.