The National Archives and Records Administration (NARA) launched the Archival Research Catalog (ARC) in 2002. ARC was the first centralized system for describing materials held throughout NARA's nationwide system of archival repositories. ARC's complicated development involved the entire agency, as staff wrestled with years of distinct descriptive…

This technical report, produced in partnership by the Council for Advancement and Support of Education (CASE) and SPSS Inc., explores the promise of data mining alumni records at educational institutions. Working with individual alumni records from The Johns Hopkins Zanvyl Krieger School of Arts and Sciences, a predictive regression model is…

This report presents the results of an experimental study which compared four different advanced retrieval methods for an online catalog to determine which are more efficient, effective, and usable. Conducted with 216 student volunteers from Virginia Polytechnic Institute and State University (Virginia Tech), the study required access to data for…

Controlled experiments were conducted with an enhanced experimental intermediary system, CONIT (COnnector for Networked Information Transfer), to test how effective such a system could be in assisting end users in online searching of medical and biomedical literature. A total of 16 end users, none of whom had previously operated CONIT or any of…

Findings are reported for a project designed to improve subject access to books by augmenting MARC records with subject descriptions. A BOOKS data base, consisting of humanities and social science monographs, was created and made available for online searching. The availability of suitable information in books to produce augmented subject…

In April 1971 the Aslib Research and Development Department began a study on selective dissemination from MARC tapes. The aim of the project was to explore the technical and economic feasibility of providing selective notifications of current books, by extraction from MARC tapes, to specialized libraries. Typical of the potential customers…

Since February 1971 the Royal Institute of Technology, Stockholm, has been running the ERIC data base mainly for SDI purposes. The implementation of the data base into the generalized search system, ABACUS, is described. One hundred and fifty-eight users received SDI service at present, 99 from governmental and educational institutions, 23 from…

Experiments studying the ability of catalog information to indicate the relevance and utility of documents by an analysis of the four main content indicating fields, e.g., the title, abstract, subject index phrases, and matching subject phrases are presented with data analysis. (Author/MBR)

The objective relevance of a document to a request is crucial to the design and testing of bibliographic retrieval systems, while subjective relevance is paramount in the use of such systems. The distinction provides a rationale for a trial-and-error mode of interaction with information systems. (Author/EM)

Describes a new Boolean retrieval environment in which an extended soft Boolean logic is used to automatically construct queries from original natural language formulations provided by users. Experimental results that compare the retrieval effectiveness of this method to conventional Boolean and vector processing are discussed. (27 references)…

Discusses the motivation for integrating information retrieval and database management systems, and proposes a probabilistic retrieval model in which records in a file may be composed of attributes (formatted data items) and descriptors (content indicators). The details and resolutions of difficulties involved in integrating such systems are…

This study investigated critical factors that would constitute reliable performance measures for bibliographic information retrieval systems, and compared the operational effectiveness of the relevance-theoretic and utility-theoretic models. The factors discussed include document and information attributes, judges, and environmental factors. (18…

Describes a document retrieval system that provides for user interaction at several search stages to acquire a detailed specification of the user's information need. The use of domain knowledge to refine the query and a browsing mechanism are described in detail, and further research questions are identified. (CLB)

A comparison of the effectiveness of information retrieval based on full-text documents with retrieval based on paragraphs, abstracts, or controlled vocabularies was accomplished using a subset of journal articles with nine search questions. It was found that full-text retrieval achieved significantly higher recall and lower precision than did the…

Examines online and printed equivalents of selected government documents indexes to compare the recall, precision, overlap, and cost effectiveness of online and manual searching. Indexes studied were the Monthly Catalog, CIS (Congressional Information Service) Index, and Government Reports Announcements and Index. (CLB)