Advances in Classification Research Volume 10. Proceedings of the 10th ASIS SIG/CR Classification Research Workshop, held at the 62nd ASIS Annual Meeting, November 1‐5, 1999, Washington, DC

Citation

Brunt, R. (2002), "Advances in Classification Research Volume 10. Proceedings of the 10th ASIS SIG/CR Classification Research Workshop, held at the 62nd ASIS Annual Meeting, November 1‐5, 1999, Washington, DC", Journal of Documentation, Vol. 58 No. 3, pp. 348-350. https://doi.org/10.1108/jd.2002.58.3.348.14

This record of the transactions at the 10th ASIS SIG/CR Classification Research Workshop presents an interesting and stimulating set of papers on a variety of themes and perspectives on classification research. Topics range from social and cultural informatics to subject access and indexing theory to modern applications of classification in electronic environments.

The book comprises eight very substantial chapters preceded by a preface, which usefully summarises the thrust of each, and rounded off with an index (which turns out to be merely to authors cited and to which we will return).

Applying a little classification we find that the papers fall into a number of groups. In one such, Andersen and Christensen (“Wittgenstein and indexing theory”) and Olson (“Cultural discourses of classification: indigenous alternatives to the tradition of Aristotle, Durkheim, and Foucault”) draw on philosophical and linguistic considerations in their quite different approaches to the nature of classification and how they might be taken into account in the fuller understanding of what we do, how we do it, and how improvement might be achieved in its practice.

Andersen and Christensen conclude that the meaning of the contents of departments must depend on the social historical and linguistic contexts in which the individual department is produced. Arguing that “the perceptions of the individual cannot and should not be incorporated” they recall, perhaps, the classified catalogue debate of the mid‐nineteenth century which challenged the logic of retrieval by subject, a debate clearly lost by those advocating purely the known item approach.

Olson discusses the cultural construction of classification through analysis of traditional schemes developed by indigenous cultures. Concluding that the underlying logic is constrained by social and cultural forces, she makes a case for working with alternative approaches. This paper is in effect continuing the debate of bibliographic imperialism manifested, for instance, in the impact of the spread of DDC to other English‐speaking countries prior even to its translation into other languages. What perhaps is of even greater import in these days of bibliographic globalisation is the fact that the MARC record carries only the great conveyors of US and modern developed Western culture. We might wait some time before the revision processes respond to her call for the paradigm shift; in the meantime, however, the analytico‐synthetic approach without fixed citation order might offer some way of reducing the impact of the rigidity of general Western‐biased schemes.

Practicalities in a number of different environments are addressed by three papers: Frâncu (“A universal classification system going through changes”), Jacob and Priss (“Nontraditional indexing structures for the management of electronic resources”) and Weedman (“Local practice and the growth of knowledge: decisions in subject access to digitized images”).

In Frâncu we hear how the challenge of harmonising UDC – a standard international scheme – with a nationally controlled vocabulary to support multilingual searching of an online catalogue was met by mapping the semantics of individual languages to the UDC structure. Much detail supported by many examples is given. However, while Derek Austin’s work on multilingual thesauri is cited, his series on PRECIS (in this journal some years ago) is not; and I wonder whether this might not have offered a more straightforward approach to providing an entry vocabulary.

Jacob and Priss revisit the field of analytico‐synthetic classification in their exploration of indexing structures for electronic resources. While describing traditional indexing in terms of DDC shelving for retrieval, it neglects Dewey’s original intention of providing a vocabulary for a systematic catalogue; and just as the potential of the classified catalogue approach of blessed memory might have been revived by Frâncu, it also offers a way to harnessing the desired discipline for the Web. Perhaps it is an indication of the vintage of this reviewer that much of what is described is not that novel: nonetheless it is gratifying to find that intellectual effort of the past is not entirely lost and may well have a bearing on retrieval in the new electronic environment.

Weedman’s paper on subject access to digitised images addresses a number of questions relating to the design and implementation of projects covering different types of classifications, concluding that subject access is the critical factor.

Two papers describing experiments in different aspects of computerised applications are found in Brooks (“Relevance auras: macro patterns and micro scatter”) and Ruiz and Srinivasan (“Combining machine learning and hierarchical indexing structures for text categorization”). Ruiz and Srinivasan compare the performances of autoclassifiers in a demonstration that the use of a hierarchical approach (based on a neural net assigning headings from MeSH to a set of Medline records) produces improved performance, indicating perhaps that human input still has a role to play. Brooks reports on experiments exploiting the relationships between semantic structures of search terms and searchers’ assessments of relevance of retrieved sets, drawing on the semantic distance model he had previously developed. The paper indicates close correlation between the data and the theoretical model, which enabled the author to suggest possible practical applications of his findings.

Uniquely, perhaps, Davenport (“Implicit orders: documentary genres and organisational practice”) reports on an interesting application of classification somewhat divorced from conventional topic specification. In her discussion of genres – which draw, almost, on the categories given in the common auxiliaries d (form) and i (points of view) of UDC – she looks at the activities of organisations and ways in which characteristics of documents produced by them and applied in their decision making may be used to analyse those operations. These different genres have, it is argued, powerful effects on individuals and groups of workers and fully justify their use in organisational studies.

All in all this is an interesting set of papers pushing at the leading edge of classification research. It shows how much that is familiar in traditional classification has considerable potential for information systems, at once removed from Dewey, Richardson, Ranganathan et al. but yet essentially close in the ultimate function of connecting searchers with the information they seek.

The one reservation concerns the index. Any publication in this subject should maintain best practices in the field and it was with some disappointment that I found that the index was only to authors cited. Disappointment was increased when the index entries, being surnames only, were found to generate confusion of citations – “Austin”, for instance, leading to two different writers. Given the wealth of knowledge in the text, a full (and conventional) back of the book index should have been provided.