论文信息 - CMU in Cross-Language Information Retrieval at NTCIR-3

CMU in Cross-Language Information Retrieval at NTCIR-3

We participated in the Cross-Language Information Retrieval evaluation at NTCIR-3 for the EnglishChinese and English-Japanese tasks. We examined several approaches to query translation, including the use of a commercial machine translation system, a thesaurus that is automatically extracted from a parallel corpus, and a general-purpose online dictionary. The MT-based approach was most effective among these alternatives in our experiments for English-Chinese retrieval on the NTCIR-2 and 3 data. Combined use of machine translation and thesaurus extraction yielded further improvement.

Yiming Yang | Nianli Ma

[1] Yiming Yang,et al. A Comparative Study on Feature Selection in Text Categorization , 1997, ICML.

[2] Kui-Lam Kwok. NTCIR-2 Chinese, Cross Language Retrieval Experiments Using PIRCS , 2001, NTCIR.

[3] Yiming Yang,et al. Translingual Information Retrieval: Learning from Bilingual Corpora , 1998, Artif. Intell..

[4] Fredric C. Gey,et al. Comparing Multiple Methods for Japanese and Japanese-English Text Retrieval , 1999, NTCIR.

[5] Yiming Yang,et al. CMU PRF using a Comparable Corpus , 2001, CLEF.

[6] C. Buckley. SMART version 11.0 , 1992 .

[7] W. Bruce Croft,et al. Phrasal translation and query expansion techniques for cross-language information retrieval , 1997, SIGIR '97.

[8] Salim Roukos,et al. Ad hoc and Multilingual Information Retrieval at IBM , 1998, TREC.

[9] Sumio Fujita. Notes on the Limits of CLIR Effectiveness: NTCIR-2 Evaluation Experiments at Justsystem , 2001, NTCIR.