W3C home > Mailing lists > Public > public-lod@w3.org > October 2018

KBpedia v 160 Now Open Source

From: Mike Bergman <mike@mkbergman.com>
Date: Thu, 25 Oct 2018 11:25:16 -0500
To: Pete Rivett <pete.rivett@adaptive.com>, "semantic-web@w3.org" <semantic-web@w3.org>, "public-lod@w3.org" <public-lod@w3.org>
Message-ID: <4ccd63d8-67ec-f1b3-24d4-a739ea483c47@mkbergman.com>
We are pleased to announce that KBpedia, a knowledge graph that 
integrates over seven leading knowledge bases including Wikidata and 
Wikipedia, is now fully available as open source. KBpedia is a 
comprehensive knowledge structure for promoting data interoperability 
and knowledge-based artificial intelligence, or KBAI. KBpedia is 
designed to support multiple use cases such as providing a computable 
framework over Wikipedia and Wikidata, creating word embedding models, 
fine-grained entity recognition and tagging, relation and sentiment 
extractors, and categorization. It is a coherent starting point for 
building your own domain knowledge graph.

KBpedia, written in RDF, SKOS and OWL 2, depending on function, includes 
55,000 reference concepts, about 30 million entities, and 5,000 
relations and properties, all organized according to about 70 modular 
typologies. Release 1.60 also has greatly increased coverage [1] over 
these leading knowledge graphs and bases:

   External source    mappings    coverage
   ---------------    --------    --------
   Wikipedia           42,108        77%
   Wikidata            27,423        50%
   schema.org             845        99%
   DBpedia ontology       764        99%
   GeoNames               918        99%
   OpenCyc             33,526        61%
   UMBEL               33,478        99%

KBpedia's upper structure, or knowledge graph, is the KBpedia Knowledge 
Ontology. We base KKO on the universal categories and knowledge 
representation guidance of the great 19th century American logician, 
polymath and scientist, Charles Sanders Peirce. The modular typologies 
are simple lists of RDF triple assertions that can be readily expanded, 
collapsed, or substituted. The thirty or so 'core' typologies are mostly 
disjoint from one another.

We subject KBpedia mappings and reference concept placements to a 
rigorous (but still fallible) suite of logic and consistency tests — and 
best practices — before acceptance. We welcome and encourage input from 
the community on gaps, errors, or inconsistencies. We desire to continue 
to grow KBpedia as a flexible and computable knowledge graph that can be 
sliced-and-diced and configured for all sorts of machine learning tasks, 
including supervised, unsupervised and deep learning.

You may explore the KBpedia knowledge graph [2], download the full 
KBpedia, KKO, mappings or typologies [3] from GitHub [4], and read 
summaries of the release on the KBpedia Web site [5] or on my blog [6]. 
All aspects of KBpedia are available under the Creative Commons 
Attribution 4.0 International (CC BY 4.0) license.

On behalf of Fred Giasson, KBpedia's co-editor, we thank you in advance 
for your feedback.

Best, Mike

[1] Coverage is the larger of percent of external concepts mapped or 
percent of KBpedia mapped to the external source
[2] http://kbpedia.org/knowledge-graph/
[3] http://kbpedia.org/resources/downloads/
[4] https://github.com/Cognonto/kbpedia
[5] http://kbpedia.org/resources/news/kbpedia-is-open-source/
[6] http://www.mkbergman.com/2168/woohoo-kbpedia-is-now-open-source/
Received on Thursday, 25 October 2018 16:25:42 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 16:22:47 UTC