Innovations in Information Retrieval: Perspectives for Theory and Practice

Citation

Sutton, V. (2012), "Innovations in Information Retrieval: Perspectives for Theory and Practice", Library Management, Vol. 33 No. 3, pp. 203-206. https://doi.org/10.1108/01435121211217289

To navigate through the massively expanding amounts of information, systems of information retrieval are becoming more and more vital, including tagging, metadata, catalogues, search engines, and classification systems. The author have considered retrieval systems for all formats of information, physical and digital as well as for new and expanding formats such video, audio, photos, web pages, and some of the complex issues that these involve.

The book covers a diverse range of topics including serendipity, semantic navigation, retrieval in fiction and music, folksonomies, social tagging, and web search engines.

Most chapters review the literature over the last 20 years, with the authors adding their perspectives to the trends. In every area researched it was illuminating how radically information retrieval has changed through the digital age so far. I would have liked more forward‐looking predictions and more ideas for future directions in this rapidly changing field. Many of the chapters concluded with a “call to action” or a hope that some of the issues and problems would be overcome.

Whilst at around 150 pages this is not a weighty tome, it is fairly heavy reading material. The text warrants careful study, and I found that to extract the most value, a second reading was useful. It is written mostly by academics and I feel the most appropriate audience would also be academics or Master's level students. Practitioners in web development, searchers and cataloguers would find it a good summary of trends in the information retrieval area over the last 20 or so years, and perhaps food for thought about where we might be going and ways to influence trends. One target audience that the book specifically states it would be useful for are students looking for possible topics for a research project, and indeed this would be a key audience as all the contributors highlighted gaps in the research and areas of possible in‐depth future study.

Between them, the eleven contributors are impressively qualified and experienced. Most are working in academic positions in schools of information studies in the UK, Europe and Canada. The authors have researched their contributions extremely thoroughly, with multiple pages of reference lists at the end of each chapter. The accuracy and proofreading throughout the whole book is faultless.

Congratulations to the editors on such a detailed and accurate index! However I also felt a glossary would be a useful addition to the book. My main criticism of the book is its inaccessibility and a large part of this is the overuse of acronyms. Not all of these were included in the index. I often found myself re‐reading back through the chapter looking for the first use with the expansion of the acronym, e.g. HCE (hit count estimates), KOS (knowledge organization system), SGs (subject gateways), m2m(machine to machine). This detracted from the flow. Even the constant use of ”IR” for information retrieval I found confusing, as even in the information field it has multiple meanings (for example, institutional repository).

As a librarian in a specialist field, I found it was interesting to read the chapters about innovations in information retrieval in diverse fields such as fiction and music, and to gain an overview of how new and innovative information retrieval options are changing information seeking behaviour. The speed of such change does raise the question of whether a book is the ideal medium for discussion of a rapidly changing field. It is almost inevitable that new innovations will have developed even before the book becomes available.

Briefly looking at the content of the book's seven chapters in more depth>

“Encountering on the road to Serendip?”, by David Bawden was my favourite chapter, which looking at browsing in new information environments. Browsing encourages creativity and innovation; however the web environment has changed the ways we can browse for information. I agree with the author, that the web does not make “unstructured serendipitous encountering” easy. He discusses how the practice and attitudes to browsing have changed and looks at ways the virtual environment can facilitate many types of browsing, but is not so useful for “goal free invitation browsing”. New forms of browsing have developed to deal with the web environment, such as rapid “flicking and bouncing” from site to site. We are challenged to understand the different ways individuals browse and to ensure our developing information systems can cater for their varied needs

Aida Slavic, who is currently the editor‐in‐chief of the Universal Decimal Classification (UDC) Consortium wrote a chapter entitled: “Classification revisited: a web of knowledge”. In this chapter she looks at how bibliographic classification, which was created for the physical arrangement of material on shelves, is evolving. She discusses the limitations and potentials of hierarchies of classification as they are used on the web in subject gateways, hubs and portals. Plenty of examples from classification systems are provided.

The chapter on fiction retrieval research was co‐authored by Anat Vernitski and Pauline Rafferty. I had no idea that so much research had been done into indexing and managing access to fiction. Vernitski describes the development of her prototype fiction retrieval system for scholars, using an intertextuality‐oriented fiction classification. The chapter covered the changes in fiction classification up to the current situation of social tagging and rating of fiction on the internet. I found myself unsettled by the authors' implications that the undisciplined user‐generated indexing initiatives using uncontrolled vocabulary were somehow less useful than a controlled system for literary scholars.

Charlie Inskip provides an illuminating chapter on music information retrieval, and discusses the challenges of classifying audio data, especially due to its lack of “word elements”. Examples of real life retrieval questions in music show how specialist approaches are required to cope with the complexities of music audio files. The author outlines some of these approaches and notes how computer‐processing power is being harnessed to search through millions of songs. Inskip's predictions of new innovations are coming to pass, as I recently read of a popular new smartphone app, SoundHound, which claims to recognise and identify songs and music played or sung to it. This illustrates the speed of innovation in this field.

User generated tagging and collaborative indexing, folksonomies and social tagging and their role in information retrieval are investigated by Isabella Peters. She looks at the growth of “recommender systems”, tag clouds, and relevance ranking and how this Web 2.0 environment creates a blurring between search and browse. The advantages and disadvantages of some of the systems are outlined. The consensus of the author and the numerous references cited is that despite its drawbacks in precision searching, this user‐generated metadata has made it possible to find huge quantities of material that would be “lost” without its tags. I like the way this chapter (unlike many of the others) illustrated her points with plenty of real‐life examples: Flickr, Amazon, LibraryThing, Delicious, etc. This immediately helped with visualising the way folksonomies work in practice.

Three authors, Richard Kopak, Luanne Freund and Heather L O'Brien worked together to produce the chapter: “Digital information interaction as semantic navigation”. They look at the way information seekers can and have engaged with and manipulated the digital resources they encounter, creating new digital content such as links, networks, annotations, blogs, reviews and tags. I found this the most interesting chapter of the book, as it shows how every information user and web searcher can enrich the content of the information for others and ways this can be encouraged. The authors discuss how “search systems can no longer afford to consider the process of retrieving information as separate to its use, as the two are inextricably linked through the user's experience”.

The final short chapter by Mike Thelwall looks at webometrics of search engines. I was looking forward to this chapter as I am always perplexed by the inconsistent and changing results that search engines return over time. Thelwall investigates research and measures on ranking, results returned over time, bias in search engine indexes and also reviews some longitudinal studies of search engines. Thelwall also points out that the search algorithms implemented by various search engines are very complex as well as commercially sensitive, so measuring results over time and difference circumstances is the best way of obtaining information on this. However, given that the search engine at the moment is perhaps the most vital access gateway to retrieving information from the vast, vague, inconsistent and misleading web, I would have liked to see more information included in this chapter on retrieving information.

Innovations in Information Retrieval provides interesting material on a fast changing dynamic topic with lots of insight by the contributing authors. I found it slightly condescending, verbose and rather too academically focussed, but detailed study of the content was worth the effort, and there are certainly many leads to follow for further research opportunities in all of the topics covered, should one feel inclined to do so.