Knowledge Graphs in AI-Driven Biomedical and Chemical Engineering: A Survey of Construction, Applications, and Future Directions

Knowledge graphs have become a popular means of storing and exploring complex data across multiple disci-plines. This survey paper explores the foundational concepts and methodologies for constructing and utilizing knowledge graphs, with a particular focus on AI-driven applications in biomedical and chemical engineering. We discuss the essential tools and frameworks commonly used in the field, followed by a detailed examination of inference algorithms tailored for these knowledge graphs. We highlight various applications, including large language models, illustrating the transformative potential of knowledge graphs in identifying drug targets, optimizing material properties, and enhancing personalized medicine. We address the challenges and limitations inherent in developing and deploying knowledge graphs, including data quality, scalability, and interoperability issues. Finally, we outline future research directions and opportunities, emphasizing the integration of advanced AI techniques and the expansion of knowledge graphs to new scientific domains.