Basic concepts of information science

ing A n abstract is 'a brief but accurate representation of the contents of a document' 122 INTRODUCTION TO INFORMATION SCIENCEINFORMATION ORGANIZATION 123 African elephant GPA DSN The African elephant is a member ofthe order Proboscidae and consistsof two distinct subspecies, the bushelephant and the forest elephantLoxodonta africanaSavanna elephant——— — _C^T Scope noteElephantsAfrican bush elephant (Loxodontaafricana africana)Forest elephant (Loxodonta africanacyclotis)Indian elephant Loxodonta africanaUSE African elephantEntry term(non-preferred term), IFigure 6.3 Example of thesaurus term (reproduced from Broughton, 2006b) (Lancaster,. 2003) or, more formally according to the relevant ISO standard 'an abbreviated, accurate representation of the contents of a document, without added interpretation or criticism and without distinction as to who wrote the abstract'.Abstracting has a long history; summaries of articles have been used as meansof keeping up with the scientific, medical and professional literature in particularfor many years. Borko and Bernier (1975) suggest that its origins can be tracedback to classical times, with the first recognizable abstracting journals, andsections for abstracts in general journals, dating from the 17th century, their useexpanding greatly in the 19th century, together with the creation of subject-specialist abstracting and indexing services such as Index Medicus and ChemicalAbstracts. For overviews, apart from the book by Borko and Bernier, seeChowdhury (2010, Chapter 8) and Koltay (2010); Alonso and Fernandez (2010)give a conceptual model for studying abstracts and abstracting.Abstracts may vary in a number of respects, and so may be categorizeddifferently. A fundamental difference is whether they are informative, intendedto give sufficient information to be a replacement for the original, or indicative,giving just enough information for a reader to decide whether the item is ofinterest. They also vary:.Abstracting has a long history; summaries of articles have been used as meansof keeping up with the scientific, medical and professional literature in particularfor many years. Borko and Bernier (1975) suggest that its origins can be tracedback to classical times, with the first recognizable abstracting journals, andsections for abstracts in general journals, dating from the 17th century, their useexpanding greatly in the 19th century, together with the creation of subject-specialist abstracting and indexing services such as Index Medicus and ChemicalAbstracts. For overviews, apart from the book by Borko and Bernier, seeChowdhury (2010, Chapter 8) and Koltay (2010); Alonso and Fernandez (2010)give a conceptual model for studying abstracts and abstracting.Abstracts may vary in a number of respects, and so may be categorizeddifferently. A fundamental difference is whether they are informative, intendedto give sufficient information to be a replacement for the original, or indicative,giving just enough information for a reader to decide whether the item is ofinterest. They also vary: • by length, from the verbose and 'literary' to the terse and 'telegraphic'• by the extent of criticism and interpretation of the original• by the extent to which they are 'targeted' or 'slanted' to a particularinterest, or type of user, as against attempting to be balanced and objective. The writing of abstracts is governed by international standard: ISO 214 (1976)Abstracts for publications and documentation'. Publishers and databaseproducers also have de facto standards; see, for example, Montesi and Owen(2007). There is a general tendency for abstracts to become more 'structured',with a consistent set of elements which are present. There still remains debateabout the most effective ways to make abstracts useful; see, for example, Hartleyand Betts (2008, 2009), Zhang and Liu (2011) and Ripple et al. (2011). This isof particular importance as more reliance is placed on abstracts because ofinformation overload (Nicholas, Huntington and Jamali, 2007). This ispotentially troubling, as studies have shown that typically 20% of abstractscontain significant inaccuracies usually presenting the subject matter of themain document in an unreasonably positive light. Automatic abstracting has been a goal for many years, since the first researchon the topic, carried out by H . P. Luhn in the late 1950s. Much effort has beenexpended since then on the design of systems, using a variety of approaches toautomatic abstracting and summarizing: see Chowdhury (2010) for an overview.Nonetheless, most practical abstracting within information systems and servicesis still largely an intellectual task. Indexing and taggingA n index is usually understood as a systematic arrangement of entries designedto enable users to locate information in a document, or in a collection ofdocuments. The process of producing such an index is 'indexing', and those whodo it are 'indexers'; it is governed by international standards ISO 5963 (1985)'Examining documents, determining their subjects and selecting