Information Representation and Retrieval in the Digital Age (ASIST Monograph Series)

The two books reviewed here cover essentially the same area: the representation and retrieval of information, in a “library/information” context. They are both aimed mainly at an LIS student audience, though one says so more clearly than the other.

Gobinda Chowdhury's book is the second edition of a well‐regarded student text, first published in 1999. The aim of the first edition was “to provide a blend of traditional and new approaches to information retrieval”. The second edition aims to maintain this approach, while incorporating “wider perspectives and latest developments”, largely those associated with the Web and with digital libraries. To a great extent, it succeeds.

The author is very clear as to his audience. The book is aimed primarily at students of librarianship and information science, both undergraduate and postgraduate. The secondary audience is practitioners who want to “brush up their knowledge in different areas of information retrieval”. Chowdhury could also have mentioned researchers, who might find it useful at the earliest stages of their studies, as a good source of literature if nothing else. But this is first and foremost a student text, and a good one.

The strengths of the book are its clarity of writing and exposition, its copious examples and references, and its breadth of coverage. It includes virtually everything that might be regarded as information retrieval, from database technology to abstracting, from bibliographic formats to natural language processing, and from models of user behaviour to mark‐up languages. This breadth, inevitably, leads to the main criticism of the book – that it does not go into depth, when depth may be called for, and that its selection of topics and examples may not always be the best. Such criticisms are easy to make, but would be somewhat unfair: the author wants to tell his readers something about everything, if it is important to information retrieval, and he mostly does it well.

Chowdhury acknowledges in his preface that the book is written “from a relatively non‐technical perspective”, suitable to its LIS student audience. Those needing the more technical approach will need to supplement this book, probably by the excellent text of Baeza‐Yates and Ribeiro‐Neto (1999), which makes a nice complement to the book under review.

The introductory chapters of the book give clear introductions to the basic concepts of retrieval and databases. Commendably, the author has expended his “cat and class” chapter of the first edition into two, to deal with bibliographic descriptions, emphasising the new MARC21 format, with the principles of cataloguing, and the relationships between AACR2 and Dublin Core, and so on. Perhaps surprisingly, there is no mention of the FRBR model, which might have provided a useful examplar of the links between bibliographic and database models.

The chapters on file organisation, vocabulary control, abstracting, etc., follow the first edition in emphasising basic principles, with a smattering of up‐to‐date examples and references. The sections on classification focus on traditional library classifications, including their Internet applications, a little more on taxonomies might have been helpful, although examples of taxonomy browsing are given in the interface chapter.

The single chapter on searching and retrieval, one of 23, seems rather limited in scope for what many readers will think to be the “core” of the subject, though it is augmented by other chapters on the specific of searching in online and CD‐ROM sources, on the Web, and in digital libraries. And, of course, Chowdhury has written separate books on all these three topics (all published by Facet), for those who need more detail.

The chapters on evaluation, building on Cranfield foundations, are enhanced by discussion of the various TREC experiments: they are well summarised, but this is one aspect in which anyone needing in‐depth material would have to look elsewhere.

The chapters on intelligent retrieval and on natural language processing again focus on the basics, but are perhaps the most disappointing aspects of the book. Intelligent retrieval is largely represented by rather dated accounts of expert system applications. There is no mention, for example, of systems like Autonomy, or ontology builders like CYC, which are now arguably significant enough to require some discussion, even to a LIS student audience.

These quibbles, however, are minor. In general, this is a very good textbook, and should be appreciated by its intended audience.

Heting Chu's book covers much the same area. It is aimed at “anyone who is interested in learning about the field, particularly those who are new to it”, and is intended to be “systematic, thorough, yet nontechnical”. I imagine that the main audience will be similar to that for the Chowdhury book – LIS students. The emphasis is on “principles and fundamentals […] rather than on descriptions of specific procedures, systems or corresponding practices in the field”, and it shows.

The author justifies “another book on information retrieval” by the need to deal with new topics and developments, as well as to deal with issues of information representation. It is – most obviously – shorter than Chowdhury, though the chapter headings show that it attempts to cover much the same scope. It does this – in keeping with the author's stated intention – by giving a much “barer” treatment, focusing on general principles without the same degree of detailed explanation, nor the same number of examples and illustrations, nor such comprehensive literature referencing. This will appeal to those who prefer a more spartan approach, and the two books are this sense complementary.

Chu begins with an overview of information representation and retrieval, including, commendably, a historical perspective. This is rather limited, however, beginning in the 1940s. Though it is true, as Chu points out, that the 1939‐45 war give the IR field an official status, the tools and techniques had been developing long before this. The pioneers who get short biographies – Taube, Luhn, Mooers and Salton – though certainly significant, are not representative of wider international developments over a longer period. Two chapters deal with information representations, focusing on subject representations and metadata, and one on retrieval languages. The next four chapters, forming the core of the book, deal with retrieval techniques, retrieval approaches including search strategies, retrieval models, and retrieval systems. They give a more concentrated and focused account than Chowdhury's book, and could be recommended as a way of appreciating the basic concepts, “unencumbered” by detail and by many examples. Multimedia retrieval, user models and interaction, and evaluation get a chapter each, and the book ends with a short account of natural language processing and intelligent systems.

Both these books are worthwhile. I suspect that Chowdhury's text will prove more popular, at least with European readers, because of the greater extent of relevant examples and context.