Michael Odetayo

BSc, MSc, DIC, PhD

Principal Lecturer

Tel: +44 02476 8253  Fax:

Email: m.o.odetayo@coventry.ac.uk

Michael Odetayo is a Principal Lecturer and the Programme Manager for IT and Joint courses in the Computer Science subject group of the School of Mathematical and Information Sciences, Coventry University, UK. Before joining Coventry University, he obtained his MSc degree from the Department of Computer Science, Imperial College, London, UK. Later he became a Chief Analyst/Programmer at the Computer Centre of Ahmadu Bello University, Zaria, Nigeria, where he led many project teams that designed and developed computer based systems for the University and other organisations outside it. He was also the course co-ordinator of the Computer Centre. After completing his PhD at the Department of Computer Science, University of Strathclyde, Glasgow, UK, he joined De Montfort University, Leicester, UK, where he was a Senior Lecturer in the Department of Computer Science for many years. He has published many papers in Genetic Algorithms and Classifier Systems. His other research areas include Expert Systems, Neural Networks, Fuzzy Logic, Data Mining and Machine Learning. He is a member of the International Programme Committee of the International Mendel Conference on Soft Computing.

Research Interests

Evolutionary systems

His current primary research area is in the field of evolutionary computation and learning systems. At the moment the systems that he is concentrating on include:

1. Hybrid/Adaptive Genetic based systems

He is interested in exploring ways of combining Genetic Algorithms with other heuristics or weak methods. Although Genetic Algorithms are general-purpose methods, they do fail sometimes to reach optimal solutions at acceptable times. However, they have some advantages over other methods (such as the ability to produce many optimal or near optimal solutions at a time) which he believes should be exploited. A hybrid system seems to be an excellent method of achieving that goal. The objective is to develop a set of problem solving heuristics that intelligently and dynamically integrate Genetic Algorithms with other similar heuristics so as to exploit and maximise their strengths.

2. Learning systems in medical applications

He is interested in employing Evolutionary Systems, Neural Networks, Fuzzy Logic systems and similar Artificial Intelligent based systems in the knowledge discovery and data mining of medical data.

External Professional Activities:

Member of the International programme committee of the International Mendel Conference on Genetic Algorithm, optimization problems, Fuzzy Logic, Neural Networks, Rough sets.

Selection of Publications:

SEKER, H., ODETAYO, M.O., PETROVIC, D., NAGUIB, R.N.G. and HAMDY, F.C. (2000) In P Sincak, J Vascak, V Kvasnicka, R Mesiar (eds): The State of the Art in Computational Intelligence, A soft measurement technique for searching significant subsets of prostate cancer prognostic markers, Physica-Verlag Heidelberg and New York. pp. 325-328 ISBN: 1615-3871

SEKER, H., ODETAYO, M.O., PETROVIC, D., NAGUIB, R.N.G. and HAMDY, F.C. (2000) Proc of the World Congress on Medical Physics and Biomedical Engineering, Ranking prostate cancer prognostic markers using a fuzzy k-nearest neighbour algorithm, Chicago, USA. 23-28 July, CD-ROM pp. 4008-33988

SEKER, H., ODETAYO, M.O., PETROVIC, D., NAGUIB, R.N.G., BARTOLI, C., ALASIO, L., LAKSHMI, M.S. and SHERBET, G.V. (2000) Proc of the 3rd IEEE EMBS Int Conf on Information Technology Applications in Biomedicine (ITAB-IT IS 2000), A fuzzy measurement-based assessment of breast cancer prognostic markers, Arlington, VA, USA, IEEE. 9-10 November pp. 174-178 ISBN: 0-7803-6449-X

Odusanya, A.A., Odetayo, M.O., Naguib, R.N.G., and Petrovic, D. A review of soft computing and gynaecological cancer. IN: Sin ák, P., and Vas ák, J. eds. Quo Vadis Computational Intelligence?, Springer-Verlag Company, 2000, pp. 485-490. (ISBN: 3790813249)

Odusanya, A.A., Odetayo, M.O., Petrovic, D., and Naguib, R.N.G. The use of evolutionary and fuzzy models in oncological prognosis. IN: John, R., and Birkenhead, R. eds. Developments in Soft Computing, Springer-Verlag, 2001, pp. 207-215.

Odetayo, M. O. "Machine Learning using a genetic-based approach, Proceedings of ESIT99 (CD-ROM format), Orthodox Academy of Crete, June 1999.

Odetayo, M. O. "Generating Rules using a Holland-based classifier learning system", proceedings of Mendel 98, 4th International Mendel Conference on Genetic Algorithm, optimization problems, Fuzzy Logic, Neural Networks, Rough sets, 1998, pages 80 85. Technical University of Brno, Brno, Czech Republic.

Odetayo, M. O. "Empirical Study of the Interdependencies of Genetic Algorithms Parameters", EUROMICRO97, proceedings of the 23rd Euromicro conference, 1997, pages 639 643. IEEE Computer Society.

Folker, Mark and Odetayo, Michael, "Conjecture Upon: The Analysis of Computer Performance by Means Evolutionary Computing", Proceedings of Computer Measurement Groups International Conference 1997, pages 425 435. Pub. Computer Measurement Group Inc.

Odetayo, M. O., "Relationship between replacement strategy and population size", Proceedings of Mendel96, 2nd International Mendel Conference on Genetic Algorithms, Brno, Czech , June 1996

Odetayo, M. O., "Structured Genetic Algorithm (SGA): A new genetic model for overcoming low viability and low variation", Proceedings of ADT Conference, 1995

Odetayo, M. O., "replacing one or two individuals at a time during reproduction: an investigation", Proceedings of Mendel95, Brno, Czech Republic, 1995

Odetayo, M. O., "Rule induction using classifier based learning method: An investigation", Proceedings of AISB Workshop on Evolutionary Computation, 1995

Odetayo, M. O., "Knowledge acquisition and adaptation: a genetic approach", Expert Systems, February 1995, Vol. 12 N0. 1

Odetayo, M.O, Dasgupta, D. "Controlling a Dynamic Physical System Using Genetic_based Learning Methods", in Practical Handbook of Genetic Algorithms New Frontier Volume II, (edited), Lance Chambers, 1995.

Odetayo, M. O. "Optimal Population Size for Genetic Algorithms: An Investigation ", IEE Colloquium on Genetic Algorithms for the Control Systems Engineering, Savoy Place, London, May 1993.