PREFACE

The second edition of the CoIL Challenge, which was held from March 17 until May 8 2000, has been a success. 147 participants registered and 43 have sent in a solutions based on a wide variety of creative approaches.

The goal of the CoIL Challenge is to promote the application of computational intelligence and learning technology to real world problems, to clarify the relations between different approaches and to stimulate the search for solutions that combine different methods. The challenge is organized by the Computational Intelligence and Learning (CoIL) cluster, which is a cooperation between four EU funded Networks of Excellence which represent the areas of neural networks (NeuroNet), fuzzy systems (ERUDIT), evolutionary computing (EvoNet) and machine learning (MLNet).

The goal this year was to predict and explain policy ownership, a direct marketing case from the insurance sector. You can read more about the problem tasks and results in the introductory paper.

The solutions that were sent in by the participants contained a wealth of valuable information on algorithms, methods and approaches to tackle real world problems like the insurance case in the challenge, which should be made available to industry and the research community. This kind of 'tacit knowledge' on how to apply computational learning technologies is hard to distill from papers that are traditionally published in research literature, which are more constrained to the 'success stories'.

Therefore we decided, only after the challenge deadline, to collect the solutions in a report. We asked the participants either permission to publish the solutions provided, or to send in a polished up extended abstract or short paper on their approach.

We hope that this is not the final word on the 2000 edition of the CoIL Challenge, but rather a starting point for further investigation. There has already been a lively discussion on the CoIL Challenge mailing list, even after the deadline. We plan to further analyze the results, in joint cooperation with the participants, and publicize the results in popular and research literature.

The only thing that remains for now is to thank all the participants, which put so much effort and creativity in their solutions and made this edition of the CoIL Challenge into a success. We are really looking forward to work with them on future challenges!
 
 
 
 

Peter van der Putten and Maarten van Someren

Amsterdam, June 2000
 
 

For more information on CoIL, visit CoILWeb: http://www.dcs.napier.ac.uk/coil/