Difference between revisions of "Augmenting Concept Definition in Gloss Vector Semantic Relatedness Measure Using Wikipedia Articles"
(Basic information on Augmenting Concept Definition in Gloss Vector Semantic Relatedness Measure Using Wikipedia Articles) |
(Wikilinks) |
||
| Line 1: | Line 1: | ||
| − | '''Augmenting Concept Definition in Gloss Vector Semantic Relatedness Measure Using Wikipedia Articles''' - scientific work related to Wikipedia quality published in 2014, written by Ahmad Pesaranghader, Ali Pesaranghader and Azadeh Rezaei. | + | '''Augmenting Concept Definition in Gloss Vector Semantic Relatedness Measure Using Wikipedia Articles''' - scientific work related to [[Wikipedia quality]] published in 2014, written by [[Ahmad Pesaranghader]], [[Ali Pesaranghader]] and [[Azadeh Rezaei]]. |
== Overview == | == Overview == | ||
| − | Semantic relatedness measures are widely used in text mining and information retrieval applications. Considering these automated measures, in this research paper authors attempt to improve Gloss Vector relatedness measure for more accurate estimation of relatedness between two given concepts. Generally, this measure, by constructing concepts definitions (Glosses) from a thesaurus, tries to find the angle between the concepts’ gloss vectors for the calculation of relatedness. Nonetheless, this definition construction task is challenging as thesauruses do not provide full coverage of expressive definitions for the particularly specialized concepts. By employing Wikipedia articles and other external resources, authors aim at augmenting these concepts’ definitions. Applying both definition types to the biomedical domain, using MEDLINE as corpus, UMLS as the default thesaurus, and a reference standard of 68 concept pairs manually rated for relatedness, authors show exploiting available resources on the Web would have positive impact on final measurement of semantic relatedness. | + | Semantic [[relatedness]] [[measures]] are widely used in text mining and [[information retrieval]] applications. Considering these automated measures, in this research paper authors attempt to improve Gloss Vector relatedness measure for more accurate estimation of relatedness between two given concepts. Generally, this measure, by constructing concepts definitions (Glosses) from a thesaurus, tries to find the angle between the concepts’ gloss vectors for the calculation of relatedness. Nonetheless, this definition construction task is challenging as thesauruses do not provide full coverage of expressive definitions for the particularly specialized concepts. By employing [[Wikipedia]] articles and other external resources, authors aim at augmenting these concepts’ definitions. Applying both definition types to the biomedical domain, using MEDLINE as corpus, UMLS as the default thesaurus, and a reference standard of 68 concept pairs manually rated for relatedness, authors show exploiting available resources on the Web would have positive impact on final measurement of semantic relatedness. |
Revision as of 08:50, 18 June 2019
Augmenting Concept Definition in Gloss Vector Semantic Relatedness Measure Using Wikipedia Articles - scientific work related to Wikipedia quality published in 2014, written by Ahmad Pesaranghader, Ali Pesaranghader and Azadeh Rezaei.
Overview
Semantic relatedness measures are widely used in text mining and information retrieval applications. Considering these automated measures, in this research paper authors attempt to improve Gloss Vector relatedness measure for more accurate estimation of relatedness between two given concepts. Generally, this measure, by constructing concepts definitions (Glosses) from a thesaurus, tries to find the angle between the concepts’ gloss vectors for the calculation of relatedness. Nonetheless, this definition construction task is challenging as thesauruses do not provide full coverage of expressive definitions for the particularly specialized concepts. By employing Wikipedia articles and other external resources, authors aim at augmenting these concepts’ definitions. Applying both definition types to the biomedical domain, using MEDLINE as corpus, UMLS as the default thesaurus, and a reference standard of 68 concept pairs manually rated for relatedness, authors show exploiting available resources on the Web would have positive impact on final measurement of semantic relatedness.
