Two stages refinement of query translation for pivot language approach to cross lingual information retrieval: A trial at CLEF 200

Kazuaki Kishida, Noriko Kando

Research output: Contribution to journalConference articlepeer-review

3 Citations (Scopus)

Abstract

This paper reports experimental results of cross-lingual information retrieval from German to Italian. The authors are concerned with CLIR in the case that available language resources are very limited. Thus transitive translation of queries using English as a pivot language was used to search Italian document collections for German queries without any direct bilingual dictionary or MT system of these two languages. In order to remove irrelevant translations produced by the transitive translation, we propose a disambiguation technique, in which two stages of refinement of query translation are executed. Basically, this refinement is based on the idea of pseudo rele-vance feedback. In the first stage, for each source query term, we select a translation candidate that appears most frequently in the set of top-ranked documents searched for a set of terms pro-vided via transitive translation of the source query. Next, in the second stage, a standard query expansion based on pseudo relevance feedback is conducted. Our experiment result showed that the two stages refinement method is able to improve significantly search performance of bilingual IR using a pivot language. However, it also turned out that performance of this method is inferior to that of machine translation method.

Original languageEnglish
JournalCEUR Workshop Proceedings
Volume1169
Publication statusPublished - 2003
Externally publishedYes
Event2003 Cross Language Evaluation Forum Workshop, CLEF 2003, co-located with the 7th European Conference on Digital Libraries, ECDL 2003 - Trondheim, Norway
Duration: 2003 Aug 212003 Aug 22

ASJC Scopus subject areas

  • General Computer Science

Fingerprint

Dive into the research topics of 'Two stages refinement of query translation for pivot language approach to cross lingual information retrieval: A trial at CLEF 200'. Together they form a unique fingerprint.

Cite this