When documentary material is indexed by a mechanized classification system, and the results judged by trained professionals, the number of documents in disagreement, after suitable adjustment, defines the error rate of the system. In a test case disagreement was 22 percent and, of this 22 percent, the computer correctly identified two-thirds of…

The logic of different types of weighting are discussed, and experiments testing weighting schemes of these types are described. The results show that one type of weighting leads to material performance improvements in quite different collection environments. (22 references) (Author)

Presents a selective review of works in 1981 concerned with subject access to information sources, with specific discussions devoted to general and theoretical works, vocabulary management, classification schemes, subject headings, indexing, and automatic indexing and classification. A 110-item bibliography is included. (JL)

A survey of automatic indexing systems and experiments has been conducted by the Research Information Center and Advisory Service on Information Processing, Information Technology Division, Institute for Applied Technology, National Bureau of Standards. Consideration is first given to indexes compiled by or with the aid of machines, including…

This report summarizes the work carried out on the automatic construction of keyword classifications and their use in information retrieval. It discusses the possible characteristics of such classifications, and potential ways of using them; and describes the various approaches to classification that were considered in terms of an overall frame of…

Index names for chemical substances have been significantly revised by Chemical Abstracts Service beginning with vol. 76. Content and rationale for these changes are described. (JB)

A proposal for a computer-aided method of building up an indexing language is made. The method involves linking terms relevant to any special retrieval system into the UDC, via a video terminal, producing a hybrid thesaurus-classification. (5 references)

Discusses the role of certain types of classifications in a modern automated environment. (Author/PF)

Reviews contributions by scholars and institutions in the fields of classification, subject headings, thesauri, indexing, abstracting, and automatic classification and indexing. Also included are works exploring the theoretical foundations of subject analysis and the history of indexing and abstracting. There are 111 references. (RAA)

Reports on a series of programs that have been developed to process data bases, consisting of textual items, and to index and arrange (classify) the data items in accordance with an automatically generated classification system. (Author)

Recounts the development of a slide storage and retrieval system at the Philip Morris Research Center in Richmond, Virginia and describes the methods used to classify and access slides in the collection. A reference list and three appendices accompany the text. (JL)

Examines the knowledge base of an intermediary expert system that provides advice about search term selection. Topics include a proposal for an integrated approach, including an option focusing on terminological attributes based on knowledge acquired from professional searchers; and the content of the knowledge base and the research needed to…

Reviews the state of the art in automatic indexing, concentrating on the literature from 1968 to 1973. (Author/PF)

This paper briefly reviews the information explosion of the last thirty years and the various attempts made to organize that information in new ways. Section B offers a brief historic review of modern classification and subject heading theory. Section C reviews the literature of automatic indexing, automatic abstracting, and automatic…

The purpose of this research is to simulate the manual process of assigning descriptors to magazine articles by machine, to investigate how successful automated assignment of descriptors to a heterogeneous text collection is when a text classifier is trained based upon a small set of manually classified example texts, and to test performance of…