Experiments with automatic indexing and a relational thesaurus in a Chinese information retrieval system

✍ Scribed by Wan, Tian-Long ;Evens, Martha ;Wan, Yeun-Wen ;Pao, Yuen-Yuan

• Publisher: John Wiley and Sons
• Year: 1997
• Tongue: English
• Weight: 295 KB
• Volume: 48
• Category: Article
• ISSN: 0002-8231
• DOI: 10.1002/(sici)1097-4571(199712)48:12<1086::aid-asi3>3.0.co;2-q

✦ Synopsis

This article describes a series of experiments with an

focused interest on applying techniques developed for interactive Chinese information retrieval system named English to collections of documents in other languages, CIRS and an interactive relational thesaurus. Two imwe hope that the problems discussed here will prove of portant issues have been explored: whether thesauri enconcern to wide sections of the information retrieval comhance the retrieval effectiveness of Chinese documents, munity. The Tipster research has involved comparisons and whether automatic indexing can compete with manual indexing in a Chinese information retrieval system.

between information retrieval approaches and results in

Recall and precision are used to measure and evaluate English and in Japanese. Other research has applied these the effectiveness of the system. Statistical analysis of techniques to Arabic information retrieval problems the recall and precision measures suggest that the use (Abu-Salem, 1992; Hmeidi, of the relational thesaurus does improve the retrieval 1995). Japanese and Arabic, while otherwise very differeffectiveness both in the automatic indexing environment and in the manual indexing environment and that ent, are both agglutinating languages with very complex automatic indexing is at least as good as manual inmorphological structure. In Arabic, subject and object dexing.

pronouns are often combined with the main verb, while prepositions and other particles are often combined with the following nouns. As a result, the type to token ratio * To whom all correspondence should be addressed. lexical-semantic relations have been observed by anthropologists in a wide range of languages (Werner &