Marrying knowledge management with knowledge graphs at KMWorld 2024

The KM and knowledge graph (KG) communities have detected each other, but so far, there has been little integration or alignment. Both communities have so much to offer each other.

The KG community has vast amounts of data and information but little knowledge. The KM world has knowledge, but most of it is disconnected from the information it was spawned from, as well as the information it is yet to create.

Dave McComb, CEO and co-founder, Semantic Arts, discussed how both can work in tandem during his KMWorld 2024 session “KM & Knowledge Graphs.”

The KG community has a well-oiled machine for organizing data and information in machine processable and beautifully visualizable formats. The KM community has perfected the ability to find and distill knowledge, but it seems to end up in text or internal wiki-like structures.

McComb introduced a framework for working together, extending each other's spheres of influence by discussing the essential difference between KM and KG, how knowledge could be stored in a graph database providing far richer avenues for access and combination, and how real-world organizations can benefit from this integration.

“If you really thought about what information meant, you would make a comprehensive model,” McComb said. “We have more silos than we’ve ever had.”

The knowledge graph community has made huge strides in how to organize and harmonize information. However, it contains no knowledge. Meanwhile, knowledge managers have figured out how to harvest knowledge.

“We have to think harder about what knowledge actually is and how to link up your knowledge with your data,” McComb said.

Knowledge graphs are based on graph databases, he explained. Key components consist of triples, which is the information, such as names, inside the graph which then links similar items that are related together.

“You want to build a system of data around things not strings” McComb said.

Knowledge graphs are backed by ontologies, which are the schema of data, information, and knowledge within the graph. Categories are where taxonomies are stored. The key to reducing complexity in information systems is by using ontologies and taxonomies in concert.

“We make up categories, which are things, and add knowledge to them,” McComb said. “Knowledge could be attached to many different things inside an ontology. Knowledge is the distillation of experience.”

Data and information are often about individual instances. But knowledge it often about what was learned over an experience, he noted.

KMWorld returned to the J.W. Marriott in Washington D.C. on November 19-21, with pre-conference workshops held on November 18.

KMWorld 2024 is a part of a unique program of five co-located conferences, which also includes Enterprise Search & Discovery, Enterprise AI World, Taxonomy Boot Camp, and Text Analytics Forum.