Entity-Linking via Graph-Distance Minimization
Entity-linking is a natural-language–processing task that consists in identifying the entities mentioned in a piece of text, linking each to an appropriate item in some knowledge base; when the knowledge base is Wikipedia, the problem comes to be known as wikification (in this case, items are wikipe...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Open Publishing Association
2014-07-01
|
Series: | Electronic Proceedings in Theoretical Computer Science |
Online Access: | http://arxiv.org/pdf/1407.7930v1 |
id |
doaj-94febc30c7a940dabd7467cff7e6e0e3 |
---|---|
record_format |
Article |
spelling |
doaj-94febc30c7a940dabd7467cff7e6e0e32020-11-24T23:25:36ZengOpen Publishing AssociationElectronic Proceedings in Theoretical Computer Science2075-21802014-07-01159Proc. GRAPHITE 2014304310.4204/EPTCS.159.4:4Entity-Linking via Graph-Distance MinimizationRoi Blanco0Paolo Boldi1Andrea Marino2 Yahoo! Research Barcelona, Spain Dipartimento di informatica, Università degli Studi di Milano Dipartimento di informatica, Università degli Studi di Milano Entity-linking is a natural-language–processing task that consists in identifying the entities mentioned in a piece of text, linking each to an appropriate item in some knowledge base; when the knowledge base is Wikipedia, the problem comes to be known as wikification (in this case, items are wikipedia articles). One instance of entity-linking can be formalized as an optimization problem on the underlying concept graph, where the quantity to be optimized is the average distance between chosen items. Inspired by this application, we define a new graph problem which is a natural variant of the Maximum Capacity Representative Set. We prove that our problem is NP-hard for general graphs; nonetheless, under some restrictive assumptions, it turns out to be solvable in linear time. For the general case, we propose two heuristics: one tries to enforce the above assumptions and another one is based on the notion of hitting distance; we show experimentally how these approaches perform with respect to some baselines on a real-world dataset.http://arxiv.org/pdf/1407.7930v1 |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Roi Blanco Paolo Boldi Andrea Marino |
spellingShingle |
Roi Blanco Paolo Boldi Andrea Marino Entity-Linking via Graph-Distance Minimization Electronic Proceedings in Theoretical Computer Science |
author_facet |
Roi Blanco Paolo Boldi Andrea Marino |
author_sort |
Roi Blanco |
title |
Entity-Linking via Graph-Distance Minimization |
title_short |
Entity-Linking via Graph-Distance Minimization |
title_full |
Entity-Linking via Graph-Distance Minimization |
title_fullStr |
Entity-Linking via Graph-Distance Minimization |
title_full_unstemmed |
Entity-Linking via Graph-Distance Minimization |
title_sort |
entity-linking via graph-distance minimization |
publisher |
Open Publishing Association |
series |
Electronic Proceedings in Theoretical Computer Science |
issn |
2075-2180 |
publishDate |
2014-07-01 |
description |
Entity-linking is a natural-language–processing task that consists in identifying the entities mentioned in a piece of text, linking each to an appropriate item in some knowledge base; when the knowledge base is Wikipedia, the problem comes to be known as wikification (in this case, items are wikipedia articles). One instance of entity-linking can be formalized as an optimization problem on the underlying concept graph, where the quantity to be optimized is the average distance between chosen items. Inspired by this application, we define a new graph problem which is a natural variant of the Maximum Capacity Representative Set. We prove that our problem is NP-hard for general graphs; nonetheless, under some restrictive assumptions, it turns out to be solvable in linear time. For the general case, we propose two heuristics: one tries to enforce the above assumptions and another one is based on the notion of hitting distance; we show experimentally how these approaches perform with respect to some baselines on a real-world dataset. |
url |
http://arxiv.org/pdf/1407.7930v1 |
work_keys_str_mv |
AT roiblanco entitylinkingviagraphdistanceminimization AT paoloboldi entitylinkingviagraphdistanceminimization AT andreamarino entitylinkingviagraphdistanceminimization |
_version_ |
1725556703389286400 |