Cross-Lingual Linking of News Stories using ESA
| Data(s) |
01/12/2012
|
|---|---|
| Resumo |
In this paper, we describe our approach for Cross-Lingual linking of Indian news stories, submitted for Cross-Lingual Indian News Story Search (CL!NSS) task at FIRE 2012. Our approach consists of two major steps, the reduction of search space by using di�erent features and ranking of the news stories according to their relatedness scores. Our approach uses Wikipedia-based Cross-Lingual Explicit Semantic Analysis (CLESA) to calculate the semantic similarity and relatedness score between two news stories in di�erent languages. We evaluate our approach on CL!NSS dataset, which consists of 50 news stories in English and 50K news stories in Hindi. |
| Formato |
application/pdf |
| Identificador | |
| Idioma(s) |
spa |
| Publicador |
Facultad de Informática (UPM) |
| Relação |
http://oa.upm.es/14458/1/02.CLINSS_DERI.pdf |
| Direitos |
(c) Editor/Autor info:eu-repo/semantics/openAccess |
| Fonte |
Proc. of CLINSS - Cross-Language Indian News Story Search | FIRE | December 2012 | Kolkata, India |
| Palavras-Chave | #Informática |
| Tipo |
info:eu-repo/semantics/conferenceObject Ponencia en Congreso o Jornada PeerReviewed |
