CORALVILLE, IA (10/23/2018) -- Today, the editors of KBpedia, a computable knowledge structure that combines seven major public knowledge bases, announced they were releasing the entire structure as open source. The complete structure includes KBpedia's upper ontology (KKO), full knowledge graph, mappings to major leading knowledge bases, and 70 logical concept groupings called typologies. The editors also announced version 1.60, with greatly expanded mappings.
KBpedia, when first released in 2016, only provided its upper portion, the KBpedia Knowledge Ontology (KKO) as open source. Michael Bergman, a KBpedia co-editor along with Frederick Giasson, said, "While we had some proprietary needs in the first years of the structure, we're really pleased to return to our roots in open source semantic technologies and software." He added, "Open source brings greater contributions and greater scrutiny, both important to growth and improvements; we are excited to continue polishing this diamond."
KBpedia is a bridge amongst seven of the leading public knowledge bases available today. KBpedia is a comprehensive knowledge structure for promoting data interoperability and knowledge-based artificial intelligence, or KBAI. KBpedia's core knowledge structure combines key aspects of Wikipedia, Wikidata, schema.org, DBpedia, GeoNames, OpenCyc, and UMBEL into an integrated whole. KBpedia's upper structure, KKO, is based on the universal categories and knowledge representation theories of the great 19th century American logician, polymath and scientist, Charles Sanders Peirce. According to KBpedia's editors, this design provides a logical and coherent underpinning to the entire structure. The design is also modular and fairly straightforward to adapt to enterprise or domain purposes.
"We began KBpedia with machine learning and AI as the driving factors," said Giasson, also the technical lead on the project. "Those remain challenging, but we are also seeing huge demands to bring a workable structure that can leverage Wikidata and Wikipedia," he said. "We are seeing the convergence of massive public data with open semantic technologies and the ideas of knowledge graphs to show the way," Giasson stated. As the ontologist of the project, though, Bergman points to Giasson's innovations in building knowledge structures as the "hidden hammer."
"It is easier to be sloppy with proprietary stuff," said Bergman. "When you pull back the curtain with open source you better have clean assignments and structure that can stand up to inspection." Bergman noted that "tens of builds" of the complete KBpedia structure were needed in the transition from the prior versions to current release. "While we have a top-down design based on Peirce, we build the entire structure from the bottom up from simple 'triples' input specifications using Fred's logical build routines," Bergman said. "We make changes, re-run the structure against logic and consistency tests, fix the issues, and run again," he added. Bergman noted the next phase in the KBpedia release plan is to release these build routines as open source.
Though tremendous strides have been made in the past decade in leveraging knowledge bases for artificial intelligence, Bergman maintains we are butting up against two limitations. "Our first problem is that we are relying on knowledge sources like Wikipedia that were never designed for AI or data integration purposes," Bergman observed. "The second problem a that we do not have repeatable building blocks that can be extended to any domain or any enterprise. AI is sexy and attractive, but way too expensive," he argued.
KBpedia splits between entities and concepts, on the one hand, and splits in predicates based on attributes, external relations, and pointers or indexes, all informed by Charles Peirce's prescient theories of knowledge representation. Bergman indicated the project would have much further to say about the project and its relation to Peirce in the coming weeks.
The new v 1.60 release of KBpedia has 55,000 reference concepts in its guiding knowledge graph, which ties into an estimated 30 million entities, most from Wikidata. The system is inherently multi-lingual, though the current release is in English only. The project hopes to see multiple language versions emerge, which should be straightforward given the dominance of links from Wikipedia and Wikidata. As it stands, the core structure of KBpedia provides direct links to millions of external reference sources.
"Finally, now with the open source release behind us, we can shift our attention to expanding the coverage of links to external sources," said Bergman. KBpedia's Web site provides links to the various open source downloads of KBpedia. The Web site also provides a working KBpedia explorer and demo of how the system may be applied to local content for tagging or analysis.
The KBpedia knowledge structure combines seven (7) public knowledge bases - Wikipedia, Wikidata, schema.org, DBpedia, GeoNames, OpenCyc, and UMBEL - into an integrated whole. These core KBs are supplemented with mappings to more than a score of additional leading vocabularies. The entire KBpedia structure is computable, meaning it can be reasoned over and logically sliced-and-diced to produce training sets and reference standards for machine learning and data interoperability. KBpedia provides a coherent overlay for retrieving and organizing Wikipedia or Wikidata content. KBpedia greatly reduces the time and effort traditionally required for knowledge-based artificial intelligence (KBAI) tasks. KBpedia was first released in October 2016 with some open source aspects, with remaining restrictions now removed. KBpedia is sponsored by Cognonto Corporation.