手法を比較

	Thesaurus Construction	Literary Warrant Analysis
分野	Library Information Science	Library Information Science
系統	Process / pipeline	Process / pipeline
提唱年	2000	1995
提唱者	Jean Aitchison, Alan Gilchrist & David Bawden; ANSI/NISO Z39.19	E. Wyndham Hulme (concept, 1911); Clare Beghtol (operationalization, 1995)
種類	Standards-based pipeline for building an information-retrieval thesaurus	Corpus-based pipeline for justifying classes and terms
原典	Aitchison, J., Gilchrist, A., & Bawden, D. (2000). Thesaurus Construction and Use: A Practical Manual (4th ed.). London: Aslib/IMI. ISBN: 9780851424460	Beghtol, C. (1995). Domain analysis, literary warrant, and consensus: The case of fiction studies. Journal of the American Society for Information Science, 46(1), 30-44.
別名	Thesaurus Building, Thesaurus Development, Controlled Vocabulary Thesaurus Design, Information Retrieval Thesaurus Construction	Warrant Analysis, Literary Warrant Study, Bibliographic Warrant Analysis, Corpus-Based Warrant Analysis
関連	3	3
概要	Thesaurus construction is the systematic building of a controlled vocabulary in which preferred terms are linked by a standardized set of relationships — equivalence, hierarchy, and association — to support consistent indexing and effective retrieval. The definitive practical methodology was set out by Jean Aitchison, Alan Gilchrist, and David Bawden in their manual Thesaurus Construction and Use, and the controlling standard in the United States is ANSI/NISO Z39.19, with ISO 25964 as its international counterpart. A thesaurus collects the terms a domain uses, resolves synonyms and homographs so each concept has one preferred label, and then wires the preferred terms together with USE/UF, BT/NT, and RT relationships plus scope notes. The result is a structured map of a subject's concepts that indexers and searchers share, reducing the mismatch between the words authors, indexers, and users choose.	Literary warrant analysis is the method of justifying the classes and terms of a knowledge-organization system by reference to the actual content of a domain's published literature, rather than to abstract logic or a designer's intuition. The principle, traceable to E. Wyndham Hulme's 1911 idea that classification should be warranted by the books that exist, was given an empirical, corpus-based operationalization by Clare Beghtol in her study of fiction studies, where she used subject descriptors in a bibliographic database to quantify how strongly the domain's literature supported particular concepts and where the field reached consensus. Within Hjørland and Albrechtsen's domain-analysis programme, literary warrant is the empirical anchor that ties a classification or thesaurus to the discourse it serves. The analysis assembles a representative corpus, extracts and counts concepts, measures their warrant, gauges consensus, and lets that evidence determine what the system should contain and how finely it should distinguish.