Difference between revisions of "A Wikipedia based Hybrid Ranking Method for Taxonomic Relation Extraction"
(Links) |
(Infobox) |
||
| Line 1: | Line 1: | ||
| + | {{Infobox work | ||
| + | | title = A Wikipedia based Hybrid Ranking Method for Taxonomic Relation Extraction | ||
| + | | date = 2013 | ||
| + | | authors = [[Xiaoshi Zhong]] | ||
| + | | doi = 10.1007/978-3-642-45068-6_29 | ||
| + | | link = https://link.springer.com/chapter/10.1007/978-3-642-45068-6_29 | ||
| + | }} | ||
'''A Wikipedia based Hybrid Ranking Method for Taxonomic Relation Extraction''' - scientific work related to [[Wikipedia quality]] published in 2013, written by [[Xiaoshi Zhong]]. | '''A Wikipedia based Hybrid Ranking Method for Taxonomic Relation Extraction''' - scientific work related to [[Wikipedia quality]] published in 2013, written by [[Xiaoshi Zhong]]. | ||
== Overview == | == Overview == | ||
This paper proposes a hybrid ranking method for taxonomic relation extraction (or select best position) in an existing taxonomy. This method is capable of effectively combining two resources, an existing taxonomy and [[Wikipedia]], in order to select a most appropriate position for a term candidate in the existing taxonomy. Previous methods mainly focus on complex inference methods to select the best position among all the possible position in the taxonomy. In contrast, algorithm, a simple but effective one, leverage two kinds of information, the expression of and the ranking information of a term candidate, to select the best position for the term candidate (the hypernym of the term candidate in the existing taxonomy). Authors conduct approach on the agricultural domain and the experimental result indicates that the performances are significantly improved. | This paper proposes a hybrid ranking method for taxonomic relation extraction (or select best position) in an existing taxonomy. This method is capable of effectively combining two resources, an existing taxonomy and [[Wikipedia]], in order to select a most appropriate position for a term candidate in the existing taxonomy. Previous methods mainly focus on complex inference methods to select the best position among all the possible position in the taxonomy. In contrast, algorithm, a simple but effective one, leverage two kinds of information, the expression of and the ranking information of a term candidate, to select the best position for the term candidate (the hypernym of the term candidate in the existing taxonomy). Authors conduct approach on the agricultural domain and the experimental result indicates that the performances are significantly improved. | ||
Revision as of 22:50, 7 October 2019
Authors | Xiaoshi Zhong |
|---|---|
Publication date | 2013 |
DOI | 10.1007/978-3-642-45068-6_29 |
Links | Original |
A Wikipedia based Hybrid Ranking Method for Taxonomic Relation Extraction - scientific work related to Wikipedia quality published in 2013, written by Xiaoshi Zhong.
Overview
This paper proposes a hybrid ranking method for taxonomic relation extraction (or select best position) in an existing taxonomy. This method is capable of effectively combining two resources, an existing taxonomy and Wikipedia, in order to select a most appropriate position for a term candidate in the existing taxonomy. Previous methods mainly focus on complex inference methods to select the best position among all the possible position in the taxonomy. In contrast, algorithm, a simple but effective one, leverage two kinds of information, the expression of and the ranking information of a term candidate, to select the best position for the term candidate (the hypernym of the term candidate in the existing taxonomy). Authors conduct approach on the agricultural domain and the experimental result indicates that the performances are significantly improved.
