dcsimg

UMBEL Brings Context and Coherence to Web Content

BURLINGTON, Mass., July 16 /PRNewswire/ -- UMBEL, a 'subject concept' (or 'topic') reference structure for the Web, was released publicly today after 12 months of development effort. Web data connected to the UMBEL structure gains context and coherence. Via UMBEL, all Web content, data, and metadata can be linked, made interoperable, and navigated and discovered more easily.

BURLINGTON, Mass., July 16 /PRNewswire/ — UMBEL, a 'subject concept' (or 'topic') reference structure for the Web, was released publicly today after 12 months of development effort. Web data connected to the UMBEL structure gains context and coherence. Via UMBEL, all Web content, data, and metadata can be linked, made interoperable, and navigated and discovered more easily.

"About a year ago we became convinced that the RDF data model and Linked Data techniques offered the first pragmatic solution to the historical 'data silo' problem of unconnected data," said Michael Bergman, co-editor of UMBEL, and CEO of Zitgist, LLC. "The only missing piece was how to bring context to that data, and so we started the UMBEL project."

UMBEL (Upper Mapping and Binding Exchange Layer) [1] is a lightweight schema (or 'ontology') for relating Web content and data to a standard set of 20,000 subject concepts. Its purpose is to provide a fixed set of common reference points in the global knowledge space. These subject concepts have defined relationships between them, and can act as semantic binding nodes for any data or Web content. The UMBEL reference structure is a large, inclusive, linked concept graph.

BASED ON THE OPENCYC KNOWLEDGE BASE

UMBEL's subject concepts and their relationships are a derived subset of the Cyc knowledge base and its open-source version, OpenCyc. Cyc has been developed over 20 years through more than 1,000 person-years of effort to capture the common sense relationships of the world.

"The UMBEL project, with its need to capture the 'open world' aspects of the Web, has brought a fresh perspective and set of challenges to Cyc," said Michael Witbrock, VP of Research at Cycorp. "We are excited by the release and committed to supporting UMBEL as part of our overall Web relevance strategy," he added. "We see UMBEL as a useful gateway to the more sophisticated reasoning and other tools surrounding Cyc."

Though there is no formal affiliation between the UMBEL project and Cyc, "the participation and responsiveness of Cycorp and The Cyc Foundation have been absolutely essential to the quality of UMBEL," said Bergman.

Cycorp and the project anticipate a series of updates and releases to both OpenCyc and UMBEL in the coming months resulting from use and community feedback.

A SEMANTIC WEB FOUNDATION

The UMBEL vocabulary defines some important new predicates and leverages existing semantic Web standards. The ontology is provided as Linked Data with Web services access (and a pending SPARQL endpoint).

Besides the 20,000 subject concepts and relationships distilled from OpenCyc, a further 1.5 million named entities have been extracted from Wikipedia and mapped to the UMBEL reference structure with cross-links to YAGO and DBpedia [2]. The system can easily be extended with additional dictionaries of named entities, including ones specific to enterprises or domains.

Fred Giasson, UMBEL's co-editor, and CTO of Zitgist, posts separately (http://fgiasson.com/blog/) on how the UMBEL vocabulary can be used to enrich existing semantic Web ontologies and techniques. Also, see the project's Web site for additional background and explanatory information on the project.

"UMBEL is a major addition to the burgeoning Linked Data Web," said Kingsley Idehen, Founder & CEO of OpenLink Software, and sponsor of the UMBEL initiative. "It provides a coherent 'data dictionary' that meshes naturally with the DBpedia project OpenLink co-founded nearly two years ago," he noted. "This data dictionary orientation of UMBEL also increases the palatability of Linked Data to the 'closed world' realm of enterprise data spaces, reducing the pain associated with the timeless quest for data integration," he added.

OpenLink has embraced UMBEL in its recently released OpenLink Data Explorer (ODE) [3]. UMBEL exploitation occurs within the Explorer's underlying RDF data access cartridges, which enable ODE to describe any (X)HTML document on the Web in ways suitable for meshing with other Linked Data.

ACCESS, DOCUMENTATION AND WEB SERVICES

UMBEL is provided as open source under the Creative Commons 3.0 Attribution-Share Alike license. The complete ontology with all subject concepts, definitions, terms and relationships can be freely downloaded.

Five volumes of technical documentation are available [4]. The two most important are the "UMBEL Ontology, Vol. A1: Technical Documentation" and "Distilling Subject Concepts from OpenCyc, Vol. B1: Overview and Methodology".

All subject concepts are Web-accessible as Linked Data URIs. A dozen various UMBEL Web services are available [5]. With this announcement, a new overview slide-show has also been posted online [6].

This release, version 0.70, is the first in which the UMBEL subject concepts and ontology will be applied as a real vocabulary in public settings. Some areas are known to be stronger and more complete than others.

Input mechanisms are being put in place for user feedback to supplement the project's discussion forum [7] and wiki [8]. The project anticipates updates based on this feedback in the coming months.

CONTRIBUTORS AND KUDOS

Zitgist, LLC (http://www.zitgist.com/) has been the major source of staff time and hosting services to the project. OpenLink Software Inc., (http://www.openlinksw.com/) has been the major source of infrastructure, financing and software for the project.

Cycorp (http://www.cyc.com/) staff have devoted many person-months to help explain the underlying system and to make improvements and revisions to OpenCyc and its OWL version in response to project input. YAGO (http://www.mpi-inf.mpg.de/~suchanek/downloads/yago/) showed how to replace the native Wikipedia structure with alternate external structures and provided the starting set of named entities used within UMBEL.

The Cyc Foundation (http://www.cycfoundation.org/blog/) and its members have been devoted to Web exposure of OpenCyc and have provided great guidance to the project in learning and navigating the knowledge base. Moritz Stefaner (http://der-mo.net/) kindly contributed UMBEL's Subject Concept Explorer and continues to make ongoing improvements to UMBEL's visualization.

REFERENCES AND LINKS [1] Project Web Site — http://www.umbel.org/ [2] DBpedia — http://dbpedia.org/ [3] OpenLink ODE — http://ode.openlinksw.com/ [4] Documentation — http://www.umbel.org/documentation.html [5] Web Services — http://umbel.zitgist.com/ [6] Slide-shows — http://www.slideshare.net/mkbergman/ [7] Discussion Forum — http://groups.google.com/group/umbel-ontology/ [8] Wiki — http://wiki.umbel.org/ Linked Data Subject Concepts and Named Entities — see [4]

SOURCE OpenLink Software Inc.; Zitgist, LLC

Comments are closed.