Inproceedings3015: Unterschied zwischen den Versionen
Pso (Diskussion | Beiträge) |
Pso (Diskussion | Beiträge) |
||
| (4 dazwischenliegende Versionen desselben Benutzers werden nicht angezeigt) | |||
| Zeile 16: | Zeile 16: | ||
}} | }} | ||
{{Inproceedings | {{Inproceedings | ||
| − | |Referiert= | + | |Referiert=True |
|Title=Cross-lingual Information Retrieval based on Multiple Indexes | |Title=Cross-lingual Information Retrieval based on Multiple Indexes | ||
|Year=2009 | |Year=2009 | ||
| Zeile 27: | Zeile 27: | ||
|Abstract=In this paper we present the technical details of the retrieval system with which | |Abstract=In this paper we present the technical details of the retrieval system with which | ||
we participated at the CLEF09 Ad-hoc TEL task. We present a retrieval approach | we participated at the CLEF09 Ad-hoc TEL task. We present a retrieval approach | ||
| − | based on multiple indexes for | + | based on multiple indexes for different languages which is combined with a conceptbased |
retrieval approach based on Explicit Semantic Analysis. In order to create the | retrieval approach based on Explicit Semantic Analysis. In order to create the | ||
| − | language- | + | language-specific indices for each language, a language detection approach is applied |
| − | as preprocessing step. We combine the | + | as preprocessing step. We combine the different indices through rank aggregation and |
| − | present our experimental results with | + | present our experimental results with different rank aggregation strategies. Our results |
show that the use of multiple indices (one for each language) does not improve upon a | show that the use of multiple indices (one for each language) does not improve upon a | ||
baseline index containing documents in all languages. The combination with concept | baseline index containing documents in all languages. The combination with concept | ||
based retrieval, however, results in better retrieval performance in some of the cases | based retrieval, however, results in better retrieval performance in some of the cases | ||
| − | considered. For the bi-lingual tasks the | + | considered. For the bi-lingual tasks the final retrieval results of our system were the |
5th best results on the BL dataset and the second best on the BNF dataset. | 5th best results on the BL dataset and the second best on the BNF dataset. | ||
| − | |Link=http://www.clef-campaign.org/2009/working_notes/ | + | |Download=Sorg-paperCLEF2009.pdf, |
| + | |Link=http://www.clef-campaign.org/2009/working_notes/ | ||
|Projekt=Multipla | |Projekt=Multipla | ||
|Forschungsgruppe=Wissensmanagement | |Forschungsgruppe=Wissensmanagement | ||
| + | }} | ||
| + | {{Forschungsgebiet Auswahl | ||
| + | |Forschungsgebiet=Information Retrieval | ||
| + | }} | ||
| + | {{Forschungsgebiet Auswahl | ||
| + | |Forschungsgebiet=Natürliche Sprachverarbeitung | ||
}} | }} | ||
Aktuelle Version vom 24. November 2009, 12:42 Uhr
Cross-lingual Information Retrieval based on Multiple Indexes
Cross-lingual Information Retrieval based on Multiple Indexes
Published: 2009
September
Buchtitel: Working Notes for the CLEF 2009 Workshop
Verlag: Cross-lingual Evaluation Forum
Erscheinungsort: Corfu, Greece
Referierte Veröffentlichung
BibTeX
Kurzfassung
In this paper we present the technical details of the retrieval system with which
we participated at the CLEF09 Ad-hoc TEL task. We present a retrieval approach
based on multiple indexes for different languages which is combined with a conceptbased
retrieval approach based on Explicit Semantic Analysis. In order to create the
language-specific indices for each language, a language detection approach is applied
as preprocessing step. We combine the different indices through rank aggregation and
present our experimental results with different rank aggregation strategies. Our results
show that the use of multiple indices (one for each language) does not improve upon a
baseline index containing documents in all languages. The combination with concept
based retrieval, however, results in better retrieval performance in some of the cases
considered. For the bi-lingual tasks the final retrieval results of our system were the
5th best results on the BL dataset and the second best on the BNF dataset.
Download: Media:Sorg-paperCLEF2009.pdf
Weitere Informationen unter: Link
Information Retrieval, Natürliche Sprachverarbeitung