Abstract:The issue of free-text versus controlled vocabulary is examined in this article. The history of the issue, which is seen as beginning with the debate over title term indexing in the last century, is reviewed and then attention is turned to questions which have not been satisfactorily addressed by previous research. The point is made that these questions need to be answered if we are to design retrieval tools, such as thesauri, upon a rational basis.
“…The expenditure of these resources is justified in the belief that controlled vocabularies, because of their classing functions, are the primary means of facilitating recall and, therefore, provide a needed service to users. 2 While authority control exists as a service to users, in practice, users have had little or no input into the construction of thesauri. With the exception of consultative groups representing specific subject fields, it is usually librarians or indexers who determine controlled vocabularies.…”
Section: Current Methods Of Con-trolling Indexing Resultsmentioning
The expanding domain within which subject searching takes place creates new challenges in providing effective subject access. Three changes in the domain of subject access are described.
“…The expenditure of these resources is justified in the belief that controlled vocabularies, because of their classing functions, are the primary means of facilitating recall and, therefore, provide a needed service to users. 2 While authority control exists as a service to users, in practice, users have had little or no input into the construction of thesauri. With the exception of consultative groups representing specific subject fields, it is usually librarians or indexers who determine controlled vocabularies.…”
Section: Current Methods Of Con-trolling Indexing Resultsmentioning
The expanding domain within which subject searching takes place creates new challenges in providing effective subject access. Three changes in the domain of subject access are described.
“…For example, Fisher and Elchesen, 1972, showed that searching title words in combination with index terms was better than searching either representation alone. Svenonius, 1986, in a review of research related to controlled vocabulary and text indexing, makes it clear that the two representations have long been thought of, and used, as complementary. A number of major studies, such as the Cranfield tests (Cleverdon, 1967), the SMART experiments (Salton, 1971), and the Cambridge experiments (Sparck Jones, 1974), also used multiple representations of documents but focused on establishing the relative effectiveness of each representation, rather than on the effectiveness of combinations of representations.…”
The combination of different text representations and search strategies has become a standard technique for improving the effectiveness of information retrieval. Combination, for example, has been studied extensively in the TREC evaluations and is the basis of the "meta-search" engines used on the Web. This paper examines the development of this technique, including both experimental results and the retrieval models that have been proposed as formal frameworks for combination. We show that combining approaches for information retrieval can be modeled as combining the outputs of multiple classifiers based on one or more representations, and that this simple model can provide explanations for many of the experimental results. We also show that this view of combination is very similar to the inference net model, and that a new approach to retrieval based on language models supports combination and can be integrated with the inference net model.
“…Durante muchos años, la recuperación a texto completo y los vocabularios controlados se han visto como enfoques alternativos al acceso por materias, cada uno con sus ventajas e inconvenientes (Walker; Janes, 1993; Taylor, 1995). Por tanto, las discusiones se pueden dividir en dos grupos: vocabulario no controlado versus controlado (Svenonius, 1986;Rowley, 1994) y el propio debate surgido dentro de los defensores del segundo grupo, es decir, precoordinado versus postcoordinado (Miller;Teitelbaum, 2002;Mann, 2003). A pesar de la controversia, hay un consenso generalizado respecto a que los vocabularios controlados y los no controlados no tienen que considerarse alternativos sino, más bien, complementarios, cada uno presenta ventajas e inconvenientes dependiendo de las situaciones de búsqueda (Kambil;Bodoff, 1998;Fugmann, 2002).…”
Section: Ficheros De Autoridades De Materiasunclassified
“…Como consecuencia, los usuarios se encuentran con una gran cantidad de problemas al realizar sus búsquedas, cuestión ésta que Ríos (1991) clasifica en dos grandes grupos: aquellos que se derivan del uso de la lógica de Boole y de los encabezamientos de materia como punto principal de acceso por temas o por contenido, y los generados por un diseño inadecuado del catálogo y por una escasa funcionalidad de la ayuda en línea. Por otro lado, la cantidad de información sobre la materia de un documento incluida en los registros marc es limitada y ha evolucionado en dos décadas desde un promedio de 1,4 encabezamientos de materia hasta 2,2, tal como demuestran los estudios realizados en el contexto internacional (O'Neill;Aluri, 1981;Svenonius;McGarry, 1993;Banks, 2003). La realidad española, y a falta de estudios concluyentes, no es precisamente mucho más halagüeña.…”
Section: Ficheros De Autoridades De Materiasunclassified
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.