Concept learning algorithm for semantic web based on the automatically searched refinement condition

Dongkyu Jeon, Wooju Kim

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

Today, the web is the huge data repository which contains excessively growing with uncountable size of data. From the view point of data, Semantic Web is the advanced version of World Wide Web, which aims machine understandable web based on the structured data. For the advent of Semantic Web, its data has been rapidly increased with various areas. In this paper, we proposed novel decision tree algorithm, which called Semantic Decision Tree, to learning the covered knowledge beyond the Semantic Web based ontology. For this purpose, we newly defined six different refinements based on the description logic constructors. Refinements are replaced the features of traditional decision tree algorithms, and these refinements are automatically searched by our proposed decision tree algorithm based on the structure information of ontology. Additional information from the ontology is also used to enhance the quality of decision tree results. Finally, we test our algorithm by solving the famous rule induction problems, and we can get perfect answers with useful decision tree results. In addition, we expect that our proposed algorithm has strong advantage to learn decision tree algorithm on complex and huge size of ontology.

Original languageEnglish
Title of host publicationSemantic Technology - Third Joint International Conference, JIST 2013, Revised Selected Papers
PublisherSpringer Verlag
Pages414-428
Number of pages15
ISBN (Print)9783319068251
DOIs
Publication statusPublished - 2014
Event3rd Joint International Semantic Technology Conference, JIST 2013 - Seoul, Korea, Republic of
Duration: 2013 Nov 282013 Nov 30

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume8388 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other3rd Joint International Semantic Technology Conference, JIST 2013
CountryKorea, Republic of
CitySeoul
Period13/11/2813/11/30

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Computer Science(all)

Fingerprint Dive into the research topics of 'Concept learning algorithm for semantic web based on the automatically searched refinement condition'. Together they form a unique fingerprint.

Cite this