A survey is given of the potential role of artificial intelligence in retrieval systems. Papers by Bush and Turing are used to introduce early ideas in the two fields and definitions for artificial intelligence and information retrieval for the purposes of this paper are given. A simple model of an information retrieval system provides a framework for subsequent discussion of artificial intelligence concepts and their applicability in information retrieval. Concepts surveyed include pattern recognition, representation, problem solving and planning, heuristics, and learning. The paper concludes with an outline of areas for further research on artificial intelligence in information retrieval systems.

Artificial Intelligence in Information Retrieval Systems.

Customized user services via interactions with leadermart

The designation of overlapping hierarchies in thesauri, first outlined in 1973, is suggested as a key element in progress towards a successful man-machine partnership. An updating, expansion and theoretical background of the 1973 proposal are given. The use of the UDC, both as a matrix and a searching aid, is postulated but is not essential. Means of distinguishing overlapping terms from other ''related terms'' are suggested, in order to make possible the accurate representation of all hirarchical relationships. At its largest, the result could be a ''Universal Reference Vocabulary'', maintained on-line only and used to construct profiles before searching via natural language and/or class numbers. It is suggested that a computer program package for a small model area within Social Sciences should be given priority.

Intelligent indexing and retrieval: A man-machine partnership

Classification of office documents is performed by almost all organizations and institutions. It is regarded as an act of identifying documents or records in accordance with a predesigned scheme or filing system. As an intellectual activity, classification includes document contents analysis and matching of its subject with the subject of some existing file. Besides the necessity of having good insight into the wide body of literature, an automatic system for classification of office documents requires considerable work related experience and practical knowledge. This type of experience and knowledge is usually case dependent and specific to a particular office, therefore creating a spectrum of different approaches. The challenge is to identify a standardized approach with a more generally applicable methodology which could be applied to various office classification systems. There are a number of techniques developed by other fields that can be successfully utilized for automatic classification. Most of them are well elaborated, tested and reported in relevant literature. Systematic analysis and some generalization of available methodology suggests that there are at least three distinct methods and groups of available techniques applicable to the classification of office documents:

/pdf/automatic-classification-of-office-documents-review-of-49dnvovvfr.pdf

Automatic classification of office documents: Review of available methods and techniques

Las cuestiones relacionadas con el control de vocabulario estan fuertemente asentadas en la literatura de la Organizacion del Conocimiento desde los origenes mismos de este campo tematico. El control de vocabulario puede ser visto como un conjunto de tecnicas y procedimientos que se aplican sobre el lenguaje para resolver problemas de comprension, ambiguedad, alcance y relacion entre terminos que expresan conceptos, y/o entre denominaciones que expresan nombres de personas, lugares, productos o instituciones. Pretende superar las dificultades de interpretacion y comunicacion propias de la utilizacion del lenguaje natural, las que son vistas como desviaciones del discurso metodico y preciso que requieren las ciencias y las disciplinas. En este trabajo se recoge la historia del termino ‘control de vocabulario’, desde su origen en el ambito de la ensenanza del ingles como una segunda lengua, a comienzos de la decada de 1930, y su pasaje al vocabulario de la Bibliotecologia y Ciencia de la Informacion hacia 1950. Se presentan en forma sistematizada cinco funciones del control de vocabulario: restriccion, desambiguacion, normalizacion, jerarquizacion y relacion. A continuacion se identifican y explican los nueve procedimientos de control de vocabulario mas utilizados. Se analizan elementos del debate historico entre quienes defienden la utilizacion de vocabularios controlados y quienes proponen el lenguaje natural como mejores herramientas para la representacion tematica del contenido de documentos y recursos de informacion, y se propugna la superacion de ese debate en virtud de las caracteristicas actuales del entorno digital en que se desenvuelve buena parte de los problemas de almacenamiento, organizacion y recuperacion de documentos y datos. Se considera la idea de que la interoperabilidad puede ser vista tambien como una modalidad particular de control de vocabulario. Se establecen conclusiones respecto a la vigencia y proyeccion del control de vocabulario.

http://revista.ibict.br/ciinf/article/download/1421/1599

Control de vocabulario: orígenes, evolución y proyección

: The role of formal, controlled vocabularies for indexing and retrieval is contrasted with the use of natural language for these activities. The following credo is advanced for large central information processors as appropriate to the 1970's. (a) Highly structured controlled vocabularies are obsolete for indexing and retrieval. (b) The natural language of scientific prose is fully adequate for indexing and retrieval. (c) Machine-aided indexing of natural language is within the state of the art. (d) Natural language retrieval can be conducted on line if the request can be stated in a phrase or a sentence.

http://www.dtic.mil/cgi-bin/GetTRDoc?AD=AD0716200&Location=U2&doc=GetTRDoc.pdf

Paul H. Klingbiel

Papers

The Future of Indexing and Retrieval Vocabularies