Stage-oe-small.jpg

Inproceedings3469: Unterschied zwischen den Versionen

Aus Aifbportal
Wechseln zu:Navigation, Suche
 
Zeile 17: Zeile 17:
 
|Month=September
 
|Month=September
 
|Booktitle=Proceedings of the ACM SIGIR International Conference on the Theory of Information Retrieval (ICTIR 2015)
 
|Booktitle=Proceedings of the ACM SIGIR International Conference on the Theory of Information Retrieval (ICTIR 2015)
|Publisher=to appear
+
|Publisher=ACM
 
}}
 
}}
 
{{Publikation Details
 
{{Publikation Details
 
|Abstract=Semantic relatedness is essential for different text processing tasks, especially in the cross-lingual setting due to the vocabulary mismatch problem. Many concept-based solutions to semantic relatedness have been proposed, which vary in the notions of concept and document representation. In our contribution, we provide a unified model that generalizes over the existing approaches to cross-lingual semantic relatedness. It shows that the main existing solutions represent different ways for constructing the concept space, which result in different document representations and implications for semantic relatedness computation. In particular, it allows us to provide theoretical justifications of existing solutions. Through the experimental evaluation, we show that the results support our theoretical findings.
 
|Abstract=Semantic relatedness is essential for different text processing tasks, especially in the cross-lingual setting due to the vocabulary mismatch problem. Many concept-based solutions to semantic relatedness have been proposed, which vary in the notions of concept and document representation. In our contribution, we provide a unified model that generalizes over the existing approaches to cross-lingual semantic relatedness. It shows that the main existing solutions represent different ways for constructing the concept space, which result in different document representations and implications for semantic relatedness computation. In particular, it allows us to provide theoretical justifications of existing solutions. Through the experimental evaluation, we show that the results support our theoretical findings.
|Download=Ictir014-zhangA.pdf,  
+
|Download=Ictir014-zhangA.pdf,
 
|Projekt=XLiMe
 
|Projekt=XLiMe
 
|Forschungsgruppe=Wissensmanagement
 
|Forschungsgruppe=Wissensmanagement
 
}}
 
}}

Aktuelle Version vom 15. August 2015, 15:36 Uhr


A Theoretical Analysis of Cross-lingual Semantic Relatedness in Vector Space Models


A Theoretical Analysis of Cross-lingual Semantic Relatedness in Vector Space Models



Published: 2015 September

Buchtitel: Proceedings of the ACM SIGIR International Conference on the Theory of Information Retrieval (ICTIR 2015)
Verlag: ACM

Referierte Veröffentlichung

BibTeX

Kurzfassung
Semantic relatedness is essential for different text processing tasks, especially in the cross-lingual setting due to the vocabulary mismatch problem. Many concept-based solutions to semantic relatedness have been proposed, which vary in the notions of concept and document representation. In our contribution, we provide a unified model that generalizes over the existing approaches to cross-lingual semantic relatedness. It shows that the main existing solutions represent different ways for constructing the concept space, which result in different document representations and implications for semantic relatedness computation. In particular, it allows us to provide theoretical justifications of existing solutions. Through the experimental evaluation, we show that the results support our theoretical findings.

Download: Media:Ictir014-zhangA.pdf

Projekt

XLiMe



Forschungsgruppe

Wissensmanagement


Forschungsgebiet