Discovery of ontologies from knowledge bases

Suryanto, Hendra; Compton, Paul

doi:10.1145/500737.500764

Cited by 29 publications

(3 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Besides the large category of text-based methods, there are a few methods developed for learning ontologies [31] from knowledge bases [22], semi-structured schemata [1], and relational schemata [11].…”

Section: Ontology Learning Methods and Toolsmentioning

confidence: 99%

Building application ontologies from descriptions of Semantic Web Services

Wang¹,

Hauswirth²,

Vitvar³

et al. 2009

Web Intelligence and Agent Systems: An International Journal

View full text Add to dashboard Cite

The existence of an increasing number of different ontologies used in semantic web service fields raises numerous interoperation and communication problems with respect to service discovery, composition, and execution. Current approaches for ontology mediation often fail due to their lack of sufficient semantic expressiveness and reasoning capability. In this paper, we present a novel approach allowing ontologies to provide self-contained semantics for service applications. We show how desired application ontologies can be generated using a new merging algorithm for service ontologies based on the names of the terms. We present experimental results and compare them to the output of the PROMPT ontology merging tool.

show abstract

Section: Ontology Learning Methods and Toolsmentioning

confidence: 99%

Building application ontologies from descriptions of Semantic Web Services

Wang¹,

Hauswirth²,

Vitvar³

et al. 2009

Web Intelligence and Agent Systems: An International Journal

View full text Add to dashboard Cite

show abstract

“…Much research has been done on creating an ontology before building an application that relies upon it [14,1], but little work exists on machine learning and knowledge acquisition to discover ontology from knowledge bases [12].…”

Section: Discussionmentioning

confidence: 99%

“…Given that we have a well-defined and domain specific knowledge base of terrorism information, we propose to investigate techniques from knowledge acquisition and machine learning disciplines to reduce or eliminate the need for a knowledge engineer. Ripple-Down Rules have had reasonable success using heuristic quantitative measures to group classes and class relations [12] and suggest that ontology discovery is possible. We plan to investigate feature construction with inductive logic programming [10], semantic networks, and procedural and rule-based knowledge representation schemes based on the Suggested Upper Merged Ontology, which provides general-purpose definitions [6].…”

Section: Discussionmentioning

confidence: 99%

Using an ontology to improve search in a terrorism database system

Gruenwald

McNutt

Mercier

14th International Workshop on Database and Expert Systems Applications, 2003. Proceedings.

View full text Add to dashboard Cite

In this paper we present our implementation of a Terrorism Database System and show how to leverage the expressive power of an ontology to improve search and document retrieval by extending the knowledge base of Oracle Text. We begin with a brief introduction to our Terrorism Database System, now in production. Then we describe the structure of the ontology, the interface for knowledge workers to add/edit/delete the content of the ontology, and the techniques for semiautomatic term generation as well as semantic content resolution when classifying documents.

show abstract

Detecting relationships among categories using text classification

Mengle

Goharian

2010

J. Am. Soc. Inf. Sci.

View full text Add to dashboard Cite

Discovering relationships among concepts and categories is crucial in various information systems. The authors' objective was to discover such relationships among document categories. Traditionally, such relationships are represented in the form of a concept hierarchy, grouping some categories under the same parent category. Although the nature of hierarchy supports the identification of categories that may share the same parent, not all of these categories have a relationship with each other-other than sharing the same parent. However, some "non-sibling"relationships exist that although are related to each other are not identified as such. The authors identify and build a relationship network (relationship-net) with categories as the vertices and relationships as the edges of this network. They demonstrate that using a relationship-net, some nonobvious category relationships are detected. Their approach capitalizes on the misclassification information generated during the process of text classification to identify potential relationships among categories and automatically generate relationship-nets. Their results demonstrate a statistically significant improvement over the current approach by up to 73% on 20 News groups 20NG, up to 68% on 17 categories in the Open Directories Project (ODP17), and more than twice on ODP46 and Special Interest Group on Information Retrieval (SIGIR) data sets. Their results also indicate that using misclassification information stemming from passage classification as opposed to document classification statistically significantly improves the results on 20NG (8%), ODP17 (5%), ODP46 (73%), and SIGIR (117%) with respect to F1 measure. By assigning weights to relationships and by performing feature selection, results are further optimized.

show abstract

Discovery of ontologies from knowledge bases

Cited by 29 publications

References 3 publications

Building application ontologies from descriptions of Semantic Web Services

Building application ontologies from descriptions of Semantic Web Services

Using an ontology to improve search in a terrorism database system

Detecting relationships among categories using text classification

Contact Info

Product

Resources

About