Virtually every company worldwide is either using or learning to use machine learning (ML) and large language models (LLMs) to improve productivity and address scalability challenges. Robotic process automation, ML, and LLMs will dramatically change the way people work. Yet, the tools lack transparency and make it difficult to understand what information was used to formulate answers, including the timeliness, reliability, IP encumbrances, and other characteristics of the data.

The most effective approach to providing metadata support to address these challenges is through well-designed ontologies and knowledge graphs as key components of the underlying data fabric. The ontologies provide structure for and across data sets, enable federated query support, and facilitate data quality, provenance, and other analyses. Yet, ontologies can be expensive to build, and require collaboration across teams inside and across organizations. The EDM Council has developed an environment and repeatable methodology to address many of the challenges in building high-quality, long-lived ontologies and knowledge graphs to support AI and other initiatives. The tools, developed over the last eight years, are open source and have been extended recently to support increasing levels of quality control, regression testing, data transformation, and continuous integration and deployment (CI/CD). Our Data Innovation Laboratory (DIL) provides hosting and process templates that have been key to the success of a number of collaborative industry projects in finance, pharma, and manufacturing.

In this talk, we will:

• describe the methodology
• provide an overview of the infrastructure
• discuss lessons learned in developing the tools we’ve built

We will also share how you can get involved and leverage the infrastructure in your own quest to use ML, LLMs, and other emerging technologies.

Speaker: Elisa Kendall

Ms. Kendall is a Partner in Thematix Partners LLC, Lead Ontologist for the EDM Council, and a graduate-level lecturer in computer science, focused on data management, data governance, knowledge representation, and decisioning systems. Her consulting practice includes business and information architecture, knowledge representation strategies, and ontology design, development, and training for clients in financial services, government, manufacturing, media, pharmaceutical, and retail domains.

Recent projects have focused on the use of ontologies to drive natural language processing (NLP), machine learning, data interoperability, large language models, and other knowledge graph-based applications. Elisa represents knowledge representation and data management concerns on the Object Management Group (OMG)’s Architecture Board, is co-editor of a number of OMG standards including the Commons Ontology Library and the Ontology Definition Metamodel (ODM), and is a contributor to a number of other ISO, W3C, and EDM Council standards, including the Financial Industry Business Ontology (FIBO) and Pistoia Alliance's Identification of Medicinal Products (IDMP) effort.