IJCNN, 2003
TUTORIAL:
Data Mining and Knowledge Discovery Using Adaptive Neural Networks
Professor Nik Kasabov, MSc, PhD
Director, Knowledge Engineering and Discovery Research Institute
Chair of Knowledge Engineering, School of IT, Auckland University of Technology (AUT)
phone: +64 9 917 9506 ; fax: +64 9 917 9501, mobile phone: +64 21 488 328
WWW http://www.kedri.info; email: nkasabov@aut.ac.nz
The tutorial gives an introduction and a practical approach to contemporary problems in data
mining and knowledge discovery when neural networks (NN) are used as a main paradigm.
The emphasis is on adaptive systems, rule extraction, and facilitating knowledge discovery.
In the first part of the tutorial the main concepts and tasks of data mining and knowledge
discovery are presented, that includes: statistical parameter analysis, dynamic data analysis and
chaos analysis, clustering, approximation, vector quantisation, regression analysis, profiling
and prototyping, classification, prediction, rule extraction. The tasks are illustrated on a list of
case studies. This part also introduces briefly the main neural network (NN) models (structures
and learning algorithms) in a form of a classification scheme. A discussion on the topic of
which NN model would be appropriate for what data mining tasks is included.
In the second part, several case studies of data mining and knowledge discovery, that fall into
different application domain areas, are presented, and demonstrated as a computer simulation,
that include: financial and economic data analysis with the use of self-organising maps;
prediction and rule extraction from time series data; gene data analysis and disease profiling
with the use of supervised evolving connectionist systems (ECOS); dynamic gene data
clustering with the use of evolving self-organising maps; data mining and pattern discovery
from EEG and ECG data; prototype learning from speech and image data with the use of ZISC
and ECOS NN.
References:
Kasabov, N., Evolving connectionist systems: Methods and Applications in Bio-informatics,
Brain Study and Intelligent Machines, Springer, London, 2002
Kasabov,N. Foundations of Neural Networks, Fuzzy Systems and Knowledge Engineering,
The MIT Press, CA, MA, 1996.
Duration: 2 hours
Prerequisite knowledge: the basics of neural networks
Professor Nikola K. Kasabov is the Director of the Knowledge Engineering
and Discovery Research Institute at the Auckland University of Technology,
New Zealand. Previously he held a position of a Professor and Personal
Chair at the University of Otago, Associate Professor at the Technical
University in Sofia, Bulgaria, and a Senior Lecturer at the University of
Essex, UK. He received MSc degrees in Computer Science and Applied
Mathematics, and a PhD degree in Mathematical Sciences from the
Technical University in Sofia. He has published over 270 works, among them 10 books, 75
journal papers, and 25 patents in the area of intelligent systems, connectionist and hybrid
connectionist systems, fuzzy systems, expert systems, speech recognition, data analysis,
bioinformatics, brain-like computing. Kasabov is a Past President of APNNA – the Asia
Pacific Neural Network Assembly, and a member of its Governing Board. He is a Senior
Member of IEEE. He is a Fellow of the Royal Society of New Zealand and a Fellow of the
NZCS. In 1991 he was awarded the RSNZ Silver Medal for Science and Technology. He was
the general chairman of series of ANNES International Conferences on Artificial Neural
Networks and Expert Systems (1993-2001), and ICONIP’97, and a co-chair of ICONIP’99 and
IJCNN’2002.