Abstract
Critical work tasks of any new domain installation are the creation of the domain knowledge graph and its population with relevant instance data. It is easier to implement and test an incremental approach. Most of the implementation effort is to conceptualize (in a knowledge graph) the structure of the new domain and to populate it with instances (data). In a proof-of-concept phase, the least-effort path is to leverage KBpedia or portions of it as is, make few changes to the knowledge graph, and populate and test local instance data. You may proceed to create the domain knowledge graph from pruning and additions to the base KBpedia structure, or from a more customized format. Some of our tasks in this area are to determine the domain and scope of the ontology; incorporate domain terminology; consider reusing existing ontologies; enumerate important terms in the ontology; define the types and the class hierarchy, especially into typologies; and define the attributes of the types. From the platform perspective, that means being able to select appropriate subsets from the knowledge base, process or transform them in some way, and then submit those result set to an external tool to conduct the designated work. Ongoing use and training demand that we adequately document all steps. If KBpedia is the starting basis for the modified domain ontology, and if incremental changes are tested for logic and consistency as they occur, then it should be possible to continue to evolve the domain knowledge graph coherently.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
Some material in this chapter was drawn from the author’s prior articles at the AI3:::Adaptive Information blog: “Open SEAS: A Framework to Transition to a Semantic Enterprise” (Mar 2010); “‘Pay as You Benefit’: A New Enterprise IT Strategy” (Jul 2010); “A Brief Survey of Ontology Development Methodologies” (Aug 2010); “A New Methodology for Building Lightweight, Domain Ontologies” (Sep 2010); “Research Shows Natural Fit between Wikipedia and Semantic Web” (Oct 2008); “Shaping Wikipedia into a Computable Knowledge Base” (Mar 2015); “Reciprocal Mapping of Knowledge Graphs” (Feb 2017).
- 2.
As Chap. 14 explains, an F1 score of 95% is still based on an annotator agreement basis of perhaps 70–80%, which means an actual F1 score error rate of, say, 65%. With ten million assertions, this translates into as many as 3.5 million being in error. If actual F1 score is at 95%, that still means 500,000 errors.
- 3.
A simple Web search of https://www.google.com/search?q=filetype:owl (OWL is the Web Ontology Language, one of the major ontology formats) shows nearly 39,000 results. Still, multiple ontology languages are available, such as RDF, RDFS, and others (though use of any of these languages does not necessarily imply that the artifact is a vocabulary or an ontology).
- 4.
Specialty search engines for ontologies include Swoogle, FalconS, Watson, Sindice, and SWSE. In addition, one can use a general search engine such as Google with a search query such as <topic> owl:equivalentClass filetype:owl. Note that the filetype might also include RDF or a variant such as N3; we can substitute other language-specific constructs of interest for owl:equivalentClass.
- 5.
OntologyDesignPatterns.org (http://ontologydesignpatterns.org/wiki/Main_Page) is a semantic Web portal dedicated to ontology design patterns (ODPs). The portal was started under the NeOn project in 2009.
- 6.
The main advantage of a grounding reference is that it allows a spoke-and-hub design for data mapping, which is tremendously more efficient than pairwise mappings. In a spoke-and-hub design, where the reference ontology is the common node at the hub, only n − 1 routes are necessary to connect all sources, meaning that it scales linearly with the number of sources and attributes. Without a grounding reference, these same mapping capabilities would require routes in a pairwise (point-to-point) approach, which also scales poorly as a quadratic function. A system of ten datasets would require n(n − 1)/2 composite mappings in the reference grounding case, but 45 in a pairwise approach. Of course, datasets themselves contain tens to thousands of attributes, compounding the map scaling problem further.
- 7.
Including, of course, explicit attempts to model intangible benefits realistically.
- 8.
James Hendler, “a little semantics goes a long way.” See http://www.cs.rpi.edu/~hendler/LittleSemanticsWeb.html.
References
L. Galárraga, G. Heitz, K. Murphy, F.M. Suchanek, Canonicalizing Open Knowledge Bases (ACM Press, New York, NY, 2014), pp. 1679–1688
N.F. Noy, D.L. McGuinness, Ontology Development 101: A Guide to Creating Your First Ontology (Knowledge Systems Laboratory, Stanford University, 2001)
M.K. Bergman, 30 Active ontology alignment tools, in AI3:::Adaptive Information. http://www.mkbergman.com/?p
O. Corcho, M. Fernandez, A. Gomez-Perez, Methodologies, tools and languages for building ontologies: Where is the meeting point? Data Knowl. Eng. 46 (2003)
D.M. Jones, T.J.M. Bench-Caponand, P.R.S. Visser, Methodologies for ontology development, in Proceedings of the IT and KNOWS Conference of the 15th FIP World Computer Congress (1998)
E.P.B. Simperl, C. Tempich, Ontology engineering: a reality check, in On the Move to Meaningful Internet Systems (Springer, Berlin, Heidelberg, 2006), pp. 836–854
E. Simperl, C. Tempich, D. Vrandečić, A methodology for ontology learning, in Frontiers in Artificial Intelligence and Applications 167 from the Proceedings of the 2008 Conference on Ontology Learning and Population: Bridging the Gap between Text and Knowledge (2008), pp. 225–249
E.P.B. Simperl, C. Tempich, Y. Sure, ONTOCOM: a cost estimation model for ontology engineering, in The Semantic Web—ISWC 2006, ed. by I. Cruz, S. Decker, D. Allemang, C. Preist, D. Schwabe, P. Mika, M. Uschold, L.M. Aroyo (Springer, Berlin, Heidelberg, 2006), pp. 625–639
E. Simperl, M. Mochol, T. Burger, Achieving maturity: The state of practice in ontology engineering in 2009. Int. J. Comput. Sci. Appl. 7, 45–65 (2010)
F. Giunchiglia, M. Marchese, I. Zaihrayeu, Encoding classifications into lightweight ontologies, in Proceedings of the 3rd European Semantic Web Conference (ESWC, 2006)
SKOS Simple Knowledge Organization System Reference: W3C Recommendation, World Wide Web Consortium (2009)
M. van Assem, V. Malaisé, A. Miles, G. Schreiber, A method to convert Thesauri to SKOS, in The Semantic Web: Research and Applications, ed. by Y. Sure, J. Domingue (Springer, Berlin, Heidelberg, 2006), 95–109
M. Poveda Villalón, Ontology Evaluation: A Pitfall-Based Approach to Ontology Diagnosis. Ph.D., Universidad Politécnica de Madrid, ETSI_Informatica (2016)
A. Halevy, M. Franklin, D. Maier, Principles of Dataspace Systems (PODS), in Proceedings of ACM Symposium on Principles of Database Systems (2006), pp. 1–9
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2018 Springer Nature Switzerland AG
About this chapter
Cite this chapter
Bergman, M.K. (2018). Building Out the System. In: A Knowledge Representation Practionary. Springer, Cham. https://doi.org/10.1007/978-3-319-98092-8_13
Download citation
DOI: https://doi.org/10.1007/978-3-319-98092-8_13
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-98091-1
Online ISBN: 978-3-319-98092-8
eBook Packages: Computer ScienceComputer Science (R0)