Effectiveness of complex index terms in information retrieval
Takenobu Tokunaga, Ogibayasi Hironori, Hozumi Tanaka · 2000
This paper explores the e#ectivenessof index terms more complex than single words in conventionalinfl(U;UVU retrieval systems. Retrieval is perfzL9( in two phases. In the first phase, a conventional retrieval method (the Okapi system) is used and in the second phase, complex index terms such as syntactic relations and single words with partof speech infM"U(#z; are introduced to rerank the resultsof the first phase. The e#ectivenessof the di#erent typesof index terms are evaluated through experiments, in which the TREC-7 test collection and 50 queries are used. The experiments show that retrieval e#ectiveness was improvedfU 32 outof 50 queries. 1 Intro ductio Indexing is a key technology ininfM""(#"; retrieval, and converts a natural language text (document) into a representation that properly describes the content of the document and can also be handled e#ciently by computers. Important propertiesof indexing are exhaustivity and specificity. Exhaustivity is a property of index descrip...