Significant progress has occurred in understanding the importance of reliable tools and approaches for managing large bodies of machine-encoded structured knowledge for well-defined problems. An open, adaptive, interoperable, distributed, high-performance approach is lacking. Adaptive models, methods, and services are lacking for Cloud implementation in master repository-like settings, where curated knowledge about subjects of interest can be produced and shared. There are countless sources and methods for compiling source data for knowledge graphs (KG), but these have issues with uniformity, quality, availability, reliability, and consistency. This project addresses deciding what information types can and/or need to be stored and shared as KGs, and devising universal ways of encoding and sharing this information. KGs contain the information we know about things. They provide means to share information about mission-critical things. We will examine first-class objects (OBI) and complex first-class phenomena: events, activities, and situations (ABI). Another challenge is determining contextualization methods. Our focus is on how to encode and exploit inter-relationships between subjects/objects, events, activities, and situations. Contextualization also necessitates methods of semantic enrichment, and using embedded model behaviors that apply graphs with runtime reasoning and logic. This is crucial to achieving context-sensitive, mission-specific operations, and to applying AI techniques to graphs.Next Generation Graph,Knowledge graphs,artificial intelligence,contextualization,object-based intelligence,activity-based intelligence,semantic enrichment,mission-specific