Modeling Recommendation and Project Deployment Automation for Semantic Knowledge Graphs

Semantic technology continues to advance but adoption still has existing and emerging challenges due to varying amounts of individual knowledge for performing development tasks, limited semantic-expert resources, as well as fast-paced time lines to produce industry products. Semantic modeling is usually time-consuming even for those with the expertise. Ad-hoc and rushed implementations may result in less ideal interoperability in downstream work, slowing down adoption of knowledge graphs in intelligent systems. The objective of this contribution is to use automation of certain expert tasks to propel semantic development efforts toward solutions that avoid compounding semantic tech debt for data alignment-based tasks and subsequent knowledge graph-based projects. This work aims to function in tandem with other existing internal data curation and deployment efforts to offer an automation-based framework and tool with results automatically deployable to a chosen project space, semantic editor, or triple store. This ultimately encourages development with artifacts that have already been assembled with all semantic dependencies to encourage using them alongside other industry standards. It also reduces expensive and redundant modeling efforts, helps to teach non-semantic experts about the implications of their modeling choices with recommendations and paradigm delineations, and eases remaining data alignment work so that results more easily follow semantic best practices.