Building concept graphs using wikipedia

Ga Hui Lee, Ki Joo Hong, Han Joon Kim

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

This paper proposes a novel way of automatically building a concept graph containing ‘ISA’ and ‘ASSO’ relationships by probabilistically analyzing connected hyperlinks within the Wikipedia articles. The concept graph can be built up by connecting four types of concept pairs: relational concept pairs, infobox concept pairs, category concept pairs and synopsis anchor concept pairs. The ‘ISA’ relationship of concept pairs can be determined by computing the subsumption probabilities between incoming links of upper concepts and outgoing links of lower concepts, which is internally represented as a partial ordering matrix. If the difference of subsumption probabilities for two concepts is small, then such a concept pair allows to define the ‘ASSO’ relationship. Our prototype system can produce a highly reasonable concept graph that contains not only noun-level concepts but also proper noun-level concepts form the Wikipedia articles. We confirm that the concept graph can be used as a knowledge base for developing various types of text applications.

Original languageEnglish
Title of host publicationApplied System Innovation - Proceedings of the International Conference on Applied System Innovation, ICASI 2015
EditorsTeen-Hang Meen, Stephen D. Prior, Artde Donald Kin-Tak Lam
PublisherCRC Press/Balkema
Pages187-191
Number of pages5
ISBN (Print)9781138028937
DOIs
StatePublished - 2016
EventInternational Conference on Applied System Innovation, ICASI 2015 - Osaka, Japan
Duration: 22 May 201527 May 2015

Publication series

NameApplied System Innovation - Proceedings of the International Conference on Applied System Innovation, ICASI 2015

Conference

ConferenceInternational Conference on Applied System Innovation, ICASI 2015
Country/TerritoryJapan
CityOsaka
Period22/05/1527/05/15

Fingerprint

Dive into the research topics of 'Building concept graphs using wikipedia'. Together they form a unique fingerprint.

Cite this