I think, therefore I classify
A Seminar and Workshop hosted jointly by ISKO UK and the BCS Information Retrieval Specialist Group (IRSG) about classification, to review how and why it is learned. on Monday 16 July, 09.00 – 18.00
Wilkes Room – British Computer Society London
1st Floor, The Davidson Building, 5 Southampton Street, London, WC2E 7HA
[Location map and directions]
From the time of Plato and Aristotle to the present day, the finest minds have been fascinated by the power of classification to make sense of the world and our existence in it. Cognitive development relies on our ability to sort things into categories, and the technique of grouping of like with like is fundamental to finding our way around the conceptual as well as the physical world. Pioneers such as Dewey, Otlet, Ranganathan and Bliss exploited this human capacity in sophisticated ways so that complex multidimensional ideas could be located in a linear sequence. More recently the proliferation of electronic networks and emergence of the Semantic Web have motivated alternative approaches to the sorting out of ideas and the linkages between them.
How do the LIS and computing professions keep pace with all that’s going on in classification? Once upon a time the principles and practice of classification dominated the curriculum for information scientists and librarians, but today they are being progressively squeezed out by other topics. Who needs classification now, if the computers are doing it all for us? And if that were true, what should the computer scientists be learning about classification?
In this seminar, we reminded ourselves of the basics: the principles of classification, the philosophy, and the research into cognitive psychology. We also heard views from five different fields where classification is applied. The employment opportunities too were considered, before we sat down in breakout groups to consider what we can and should be doing about developing our understanding and skills, and how we engage the next generation of knowledge organizers. During refreshment breaks the sponsors of this event demonstrated software designed to handle classification automatically.
Where they are available, links have been provided
for each item in the programme as follows:
or slide presentation
sound recording (the remaining ones will be added when sound recordings have been edited)
text of full paper or notes
SPEAKERS AND ABSTRACTS
Vanda Broughton, Department
of Information Studies, University College
Principles of classification for libraries physical or virtual
To set the scene for this seminar, Vanda described the continuing relevance of classification for the library and information sector, what the conceptual basis is, and what an understanding of principles offers to theorists and practitioners alike.
Vanda Broughton is a Senior Lecturer in the Department of Information Studies at University College London (UCL), where her research interests are centred on faceted classification and its application, particularly in digital environments. Vanda is the author of the Facet Publishing book “Essential classification”. She is also Editor of the “Bliss Bibliographic Classification” 2nd edition and Associate Editor of the “Universal Decimal Classification”. Much of her ongoing research is into the development of faceted terminologies for various subject domains in both of these schemes.
How is classification currently taught? Review of current courses in UK
The curriculum content of computer science, library and information management degree programmes in the UK is surveyed to assess what is taught about classification and categorization, and how that knowledge is applied. The findings were compared with some recent surveys of curricula in other countries.
Christine Urquhart is a Senior Lecturer at Aberystwyth University. Her research interests are mainly in information behaviour and health information management. Her teaching in systems and business process analysis has provided a different perspective to the normal library thinking on the principles of classification as did her Cochrane systematic reviewing work on nursing record systems.
Donald Lickley, Sue Hill
Demand for classification skills: the employment/recruitment trends
What new roles have we seen in the last year? What skills are in most demand? How best to prepare oneself for the employment market?
Donald joined Sue Hill Recruitment in January 2011 and has over 20 years' experience in the information world. He has a background in the management and delivery of HR support and advice, and staff development and training gained within Higher Education. He also has frontline experience of working in and managing information and library services in a variety of sectors, from Lambeth Palace Library to HM Treasury to King's College London, and various points between and beyond.
Sandra Knapp, Department of
Botany, Natural History Museum
Order out of chaos – classification and naming in biology
Human beings have long classified the world around them; the creation myths of most religions have as significant elements a classification of the rest of life on Earth. Medieval herbalists recognised groups of plants by their uses, and gradually systems for the classification and naming of organisms became essential for reference and learning purposes, but increasingly complex and unwieldy. Carl Linnaeus in the 18th century and Charles Darwin in the 19th century both provided concepts leading to paradigm shifts in naming and classification. Cladistic tree-thinking beginning in the mid-20th century did the same. I discussed the role of biological classification in the study of biology itself, and how the intertwining of classification and naming can cause problems for biologists. Most biology today is done with reference to the phylogenetic or evolutionary Tree of Life - this may or may not be a classification in the strict sense. Do we even need a classification of life on Earth today, or will the tree itself be sufficient?
Dr Sandra Knapp is a graduate of Pomona College in California and of Cornell University, Ithaca, New York. She is a specialist in the taxonomy of the nightshade family, Solanaceae, and has spent much time in the field in Central and South America collecting plants. Her particular focus of research is the taxonomy of Solanum, the megadiverse genus that contains potatoes, tomatoes and eggplants. She came to the Natural History Museum, London, in 1992 to manage the international project Flora Mesoamericana – a synoptic inventory of the approximately 18,000 species of plants of southern Mexico and the isthmus of Central America. She is also the author of several popular books on the history of science and botanical exploration, including the award-winning “Potted Histories” (2004). She is actively involved in promoting the role of taxonomy worldwide. In 2009 she won the Peter Raven Outreach Award from the American Society of Plant Taxonomists and the UK National Biodiversity Network’s John Burnett Medal.
Institute for the Science and Technologies of
Information, Italian National Research Council
Recent advances in automatic classification
This talk addressed recent advances in automatic classification brought about by research at the crossroads of information retrieval, machine learning, and computational linguistics.
Fabrizio Sebastiani is a Senior Researcher at the Institute for the Science and Technologies of Information of the Italian National Research Council, where he leads the Text Mining Group. On the theme of automatic classification he has published several papers in international journals and at international conferences, and he has taught tutorial courses at international summer schools, international conferences, and PhD programs.
Judi Vernau, Metataxis
Classification functionality in search and browse
People look for information in different ways, and so need different semantic structures to support them. Or do they? Can one ‘size’ of classification scheme fit all, and are the traditional structures associated with them still relevant?
Judi Vernau is an Information Architect working for Metataxis, a consultancy she co-founded in 2002. She started her information career as a librarian before moving into publishing, where she worked on structuring data and developing classification schemes for some of the earliest electronic reference products. After working for Reed Elsevier, Macmillan and EMAP, where information architecture was focused entirely on published products, she expanded her knowledge and experience of the wider field of information management courtesy of The Stationery Office, where she worked with a variety of clients in the public and private sectors. At Metataxis she specializes in content analysis and structuring, metadata and taxonomy.
Steve Bailey, Senior Adviser,
Classification in records management: part of the solution or part of the problem?
Record management orthodoxy has functional classification at its core, as is reflected in professional standards, training and literature. As a result you might expect records managers to be at the forefront of thinking and technology in this area. You might also expect records managers to be exploiting functional classification as a means of integrating records management controls within the rest of the business but neither of these seem to be the case in practice. In this session we explored why this gulf between theory and practice exists and what its implications might be – both for records management and for the organisations and users it strives to serve. We also touched on whether functional analysis is a goal worth striving for, or whether new approaches to classification based around folksonomies and tagging might offer richer pickings for the records manager.
Steve currently acts as senior adviser on records and information management issues for JISC infoNet, an advisory service for managers within the HE and FE sectors. He is responsible for preparing and disseminating a range of guidance material and tools to help support the development of records and information management within the sector, including the widely acclaimed Impact Calculator. In 2011 Steve was a shortlisted finalist for SLA Europe’s ‘Information Manager of the Year award’ and in 2012 won the Emerald Literati Network Award for ‘Outstanding Paper of the Year’ for his paper “Measuring the impact of records management: data and discussion from the UK higher education sector”. Steve is the author of “Managing the crowd: rethinking records management for the Web2.0 world”, available from Facet Publishing. See also his blog at http://rmfuturewatch.blogspot.com/.
Ian Horrocks, Oxford
University Department of Computer Science
Logical foundations for a semantic web
Semantic Web research aims to evolve the World Wide Web into a platform for “intelligent” applications that can exploit a huge network of shared information. Ontologies play a key role by allowing applications to share not only data, but knowledge about the meaning of data. The logic based ontology languages and reasoning systems developed for this purpose are profoundly influencing the development of modern information systems.
Ian Horrocks is a Professor in the Oxford University Department of Computer Science where he leads the Information Systems Group. He was one of the key authors of the OIL, DAML+OIL, and OWL ontology language standards, and chairs the W3C working group that has just completed the standardisation of OWL 2. He is a Fellow of the Royal Society, an ECCAI Fellow, and is amongst the most highly cited authors in Computer Science.
Département de psychologie,
Université du Québec à
To cognize is to categorize: from the ‘blooming, buzzing confusion’ to the ‘cognitive commons’
A category is a kind of thing (object, event, action, property, state). To categorize is to do the right thing with the right kind of thing. Hence most of human and animal cognition and adaptive behaviour is categorization. The feature detectors for some categories are inborn, but most categories need to be learned. There are two ways to learn categories: by trial-and-error induction, with error-correcting feedback from the consequences of correct and incorrect categorization – or by verbal instruction (language). Our species is the only one that can learn categories through verbal description and definition. That is the main evolutionary advantage conferred by language. Our dictionaries, encyclopaedias and, lately, all of Google space contain names of our categories and their definitions/descriptions, our “category commons”. New categories can be learned from descriptions combining the names of already learned categories into a subject/predicate proposition. How many words are needed to ground an entire dictionary, or all of Google space?
Stevan Harnad was born in Budapest, Hungary. He did his undergraduate work at McGill University and his graduate work at Princeton University’s Department of Psychology. He is currently Canada Research Chair in cognitive sciences and Professor of Psychology at Université du Québec à Montréal (UQAM) and adjunct professor of cognitive science at the University of Southampton. He was elected external member of the Hungarian Academy of Sciences in 2001. His research is on categorization, communication, cognition and consciousness and he has written extensively on categorical perception, symbol grounding, origin of language, lateralization, the Turing test, distributed cognition, scientometrics, and consciousness.
John Dupré, Professor
of Philosophy of Science, University of Exeter
Philosophy of classification
Classification has been a problem at the heart of philosophy since Plato deployed his theory of forms to explain what made two things things of the same kind. Ideas about species and genera, and an account of the essence of a kind, worked out by Aristotle continue to exert an influence on contemporary philosophical thought. However biological classification, which has always been the engine-room of philosophical thought about kinds, has been radically transformed, first by Darwinism, and much more recently by genomic phylogeny, and the insights from these fields have brought an end even to modest Aristotelian essentialism. In its place increasing numbers of philosophers are accepting a radically pluralistic view of kinds as often overlapping and cross-cutting; at the same time there is a growing appreciation among philosophers of the importance of the truism that classifications are human constructs for human purposes.
John Dupré is a philosopher of science whose work has focused especially on issues in biology. He is currently Professor of Philosophy of Science at the University of Exeter and since 2002 he has been Director of the ESRC Centre for Genomics in Society (Egenis). He has formerly held posts at Oxford, Stanford, and Birkbeck College, London, and in 2006 he held the Spinoza Visiting Professorship at the University of Amsterdam. He is President of the British Society for the Philosophy of Science and a Fellow of the American Association for the Advancement of Science. His most recent book is Processes of Life: Essays on the Philosophy of Biology (Oxford 2012).
This meeting was fully booked. We're sorry if you wanted to come, but there were no places left. We have made slide presentations available on this site and will add sound recordings of the event as soon as they have been edited and prepared.
You might be interested in the next ISKO UK event, "The shape of knowledge" - information about this and a booking form are at http://www.iskouk.org/events/shape_of_knowledge_Sep2012.htm.