Airport buildings can significantly impact the environment throughout their life cycle. With over 40,000 existing numbers worldwide, airport buildings pose a serious threat that is unacknowledged either in the built environment or the aviation sector. This gap along with the lack of specific frameworks highlights the need for more airport-based research on green building performance assessments and the development of airport-specific green building rating tools (GBRT). This research aims to develop a decision tree-based modeling approach for evaluating the green performance of airport buildings. Information from publicly available airport communications such as sustainability, environment, corporate social responsibility, and annual reports are utilized to create an environmental dataset with 17 environmental features. Structured web-mining and content analysis approaches are followed to build the dataset comprising 577 report information. The Classification and Regression Tree (CART) model is preferred for the supervised learning exercise due to its high-performing ability with scarce data and easier interpretability of its results by airport stakeholders. The CART model resulted in the development of 49 green rules that are of practical use to airport operators towards their built environment management relating to the environmental categories of GBRT. The developed CART model exhibits useful and justifiable inter-relationships between features that are previously unobserved in any GBRTs.