Knowledge Graphs as Models of Integrated Digital Environments
Jaya Kambhampaty, Jorge L. Ortiz Solano, Hunter Strauss, Arun Palaniappan, Steven Bisso, Olivia J. Pinon-Fischer, Dimitri N. Mavris, John F. Matlik, Alexander H. Karl, Jonas Dahlstrom · 2024
Innovative techniques are required to manage and architect data environments at scale for large enterprises. As Artificial Intelligence and Machine Learning (AI/ML) accelerate data production in many contexts, similarly advanced methods will be required to parse, organize, and analyze that data. This paper discusses a methodology and minimum viable product pipeline to do so with metadata and a graph database, addressing the influence of metadata type and metadata-to-graph conversion schema. The graph database is used to visualize connections within the data, especially as it is influenced by the rules used to represent the metadata artifacts of the graph. Several measures of the influence of these architectural decisions are presented alongside discussion of techniques that increase the ability of generating the knowledge graph at scale.