Search results

This article reviews the state of the art in automatic indexing, that is, automatic techniques for analysing and characterising documents, for manipulating their descriptions in searching, and for generating the index language used for these purposes. It concentrates on the literature from 1968 to 1973. Section I defines the topic and its context. Sections II and III consider work in syntax and semantics respectively in detail. Section IV comments on ‘indirect’ indexing. Section V briefly surveys operating mechanized systems. In Section VI major experiments in automatic indexing are reviewed, and Section VII attempts an overall conclusion on the current state of automatic indexing techniques.

Questions put to Mr Aitchison during the evening dealt mainly with the method of his SDI investigation. In reply to Mr C. S. Sabel (Central Electricity Generating Board) Mr Aitchison said that progressive modification of user profiles would be attempted. In answer to a point made by Mr D. V. Arnold (ICI Heavy Organic Chemicals Division) he affirmed his belief that co‐operation with librarians was essential. Mr P.E.Colinese (Central Electricity Generating Board) asked about the method of assessment of relevance by users. Mr Aitchison indicated that this would become more stringent as the study progressed. In reply to Mr D.H. Barlow (McGraw Hill) he said that three indexers would be required for indexing a maximum of 15,000 documents a year. To questions on the amount of computer time required and the percentage of key word match which would enable him to regard a document as relevant to a user, he replied that they hoped to have answers within the next four months. Mr C. S. Sabel (Central Electricity Generating Board) wondered what was the best way for work on mechanization to be co‐ordinated. Mr Aitchison thought that since so few people were working in this field there was little risk of fruitless duplication of effort.

Mountbatten offers a vivid description of the current‐awareness function using the analogy of a very wide conveyor‐belt, representing the information publishers, on which books, periodicals and reports appear at random: ‘The searcher is on a platform just above the belt and as the information material passes underneath he can pick up and read anything that he thinks might be of interest to him. You can imagine his frustration as he realises that for every item he takes time to examine, hundreds of others of possible interest to him have passed by’. Personality and environment will determine whether the individual can find an intelligent compromise between the extremes of neurosis induced by worrying about the material he is missing, or complacency with any system which produces one or two interesting items.

In June 1959 a questionnaire, which is printed on page 400, was sent to all members of the Aslib Aeronautical Group other than overseas members, and public, university and (general) college libraries. Of the sixty‐nine questionnaires sent out, replies to forty‐nine were received.

Downloading and uploading offer labour‐saving advantages and are now accepted as useful options in online searching. All aspects are here considered, from recent technical advances, applications and legal attitudes. There is also a review of current software for downloading. Recent developments mean a trend to higher internal memory and storage capacity, and greater transmission speeds. Packages now offer access to more than one host, give maximum assistance to the user without being menu‐driven and incorporate the latest developments in artificial intelligence. Disadvantages are in the length of time involved in the process and the fact that the legal issue of copyright has not yet been finalised. Database producers have turned to licensing under contract law, but there is still need to rely on user ethics, and the need for a standard permissions form is highlighted.

The performance of a retrieval system with a file of only a few hundred references can be measured by assessing the relevance of each reference to each of a number of queries. A suitable measure of retrieval performance is then the Recall ratio—the fraction of the relevant references that are retrieved by the system. When the file is large this method of measuring performance is not practicable, and Recall cannot be measured although it can be estimated. A number of estimation procedures are examined and found unsatisfactory. Another measure of retrieval performance is the Extension ratio which is approximately the ratio of the quantity of known relevant references before and after putting a query to the retrieval system. The properties of this measure are examined and it is applied to 315 MEDLARS searches.

From investigations made by the NERC Working Party on Scientific Information it is clear that, even in the restricted field of electronics research, there are already too many different sources of current‐awareness information, mainly abstracts journals and title lists, which must be scanned by a research worker who wishes to keep abreast of developments in his field. To ensure that he knows of all the published and semi‐published information of interest to him he may have to scan more than ten current‐awareness publications each month.

Results of an in‐depth study of the electronic publishing (EP) industry, with particular emphasis on the consumer marketplace, are presented. EP was defined as the use of electronic media to deliver information to users in electronic form or from electronic sources. EP is contrasted to electronic‐aided publishing, which is the use of electronic means to format and produce a conventional information product. An “information chain” model of the information flows between publishers (or producers) and users was helpful in understanding the boundaries of EP and defining its markets. Following a review of the conventional publishing industry, a model of the forces driving the EP industry was derived. Although technology is the strongest driving force, it is by no means the only one; the others are economics, demographics, social trends, government policies, applications growth, and industry trends. Each of these forces is described in detail in a “cause and effect” scenario, from which keys to success in the EP marketplace are derived. Although there is some turmoil in the industry, with new services continuing to appear and disappear, the overall picture is one of optimism. EP should be a significant part of consumers' lives by the end of the decade.

Sir Raymond Streat, C.B.E., Director of The Cotton Board, Manchester, accompanied by Lady Streat. A Vice‐President: F. C. Francis, M.A., F.S.A., Keeper of the Department of Printed Books, British Museum. Honorary Treasurer: J. E. Wright, Institution of Electrical Engineers. Honorary Secretary: Mrs. J. Lancaster‐Jones, B.Sc., Science Librarian, British Council. Chairman of Council: Miss Barbara Kyle, Research Worker, Social Sciences Documentation. Director: Leslie Wilson, M.A.

IT would be quite impossible adequately to report a Dublin conference of any kind in purely professional terms. The warm friendliness of its people demands an equally personal reaction from its visitors and for public librarians certainly this is as it should be, because we are ourselves, above all, involved with people. So professional affairs at this conference were kept in their proper place—as only a part of the whole and merely providing a framework round which the business of renewing contacts and making friends could take place.