Research Seminar 'DIGITAL OBESITY addressed by Fuzzy Information Management' by Prof T. Martin (Bristol University and BT)
Wednesday 02 July 2008, 1300-1400
The phrase "digital obesity" summarises a range of problems arising from our propensity to generate and retain a rapidly growing volume of data, at web-scale as well as at corporate and personal scales. Much of this data is effectively wasted unless we can find and use the "right" data when needed.
Statistical methods help to a degree with numerical data, although they sometimes "average out" useful information. Textual data is less amenable to formal treatment. To a large degree, the problem arises from a mis-match between the precisely defined terms used by formal models of data storage and retrieval and the far more subtle and expressive terms used in human communication. We are adept at communicating in a language where the majority of concepts are fuzzy, defined by common usage rather than by necessary and sufficient conditions. It is only when we interact with computers that we adapt our way of thinking to artificial precision. We need to redress this balance, so that computers adapt to our way of thinking.
The success of fuzzy control is one example of the way in which fuzzy set theory enables computers to work with commonly understood terms such as "hot" and "slow" rather than with precise numerical values.
This talk describes some ways in which fuzziness enables computers to work with ill-defined concepts, leading to more effective use of text- based information in business and other situations.
Trevor Martin is Professor of Artificial Intelligence at the University of Bristol. Since 2001 he has been funded by BT as a Senior Research Fellow, researching soft computing in intelligent information management including areas such as the semantic web, soft concept hierarchies and user modelling. In addition, he has investigated the use of intelligent data analysis for extracting information from home sensor networks.
He is a member of the editorial board of Fuzzy Sets and Systems, and has served on many conference programme and organising committees, including programme chair for the 2007 IEEE Fuzzy Systems Conference.
He has published over 200 papers in refereed conferences, journals and books, and is a Chartered Engineer and member of the BCS and IEEE