Dr Ken Adams, Mathematics Lecturer of the School of Engineering, recently had his research paper “Simple Heuristics for the Choquet Integral Classifier” published in the Second International Conference on Electrical, Electronics, Computer Engineering and their Applications which held by The Society of Digital and Wireless communications (SDIWC) at Manila, Philippines from Feb. 11th to 14th, 2015. Up to 200 delegates from about ten different countries attended the conference to discuss and exchange research findings on computer science and technology, and new algorithms.
SDIWC, an organization devoted to promote science and technology in the digital world, hosts multidisciplinary academic conferences primarily in the fields of computer software and hardware, networking, and wireless communication. It also encourages and supports international collaborative work by involving many bilateral or multilateral cooperative projects.
Dr Adams’ research is mainly about data mining which is to extract and analyze information from data by using computer. And in this paper, Dr Adams developed heuristics based on the Choquet Integral which is a well-known and successful classification method. When the Choquet Integral is applied to to large sets of data, many coefficients need to be optimized which need search method like genetic algorithms. And the fresh idea on heuristics elaborated in Dr Adams’ paper enables optimization firstly using a smaller set of coefficients with reasonably accurate results from the data set.
The Wisconsin Breast Cancer data set is used in this research. According to Dr Adams, his paper shows that simple heuristic techniques found suitable coefficients for the design of a Choquet Integral classifier that then correctly predict who will develop cancer for more than 95% of the patient records on the data set. This will make a significant contribution to the breast cancer treatment. Now modern techniques take different types of measurements, for example genetic profiles, and further research has to be completed to see how effective the new heuristics will be on new data. As in most other countries, breast cancer is now the most common cancer in Chinese women; cases in China account for 12·2% of all newly diagnosed breast cancers and 9·6% of all deaths from breast cancer worldwide. (1). in 2008, GLOBOCAN reported breast cancer to be the sixth leading cause of death from cancer in Chinese women (2). About his research, Dr Adams said, “an accurate early diagnosis will identify those patients who should have the little developing tumor removed. This may save the lives of many patients quickly. Other methods can have similar accuracy but they are generally more complicated.” (Dr Adams paper is available at http://sdiwc.net/digital-library/simple-heuristics-for-the-choquet-integral-classifier.html )
Lei Fan, Kathrin Strasser-Weippl, Jun-Jie Li, Jessica St Louis, Dianne M Finkelstein, Ke-Da Yu, Wan-Qing Chen, Zhi-Ming Shao, Paul E Goss, (2000), “Breast cancer in China,” The Lancet Oncology,Vol.15, June 2014, pp.279
Lei Fan, Kathrin Strasser-Weippl, Jun-Jie Li, Jessica St Louis, Dianne M Finkelstein, Ke-Da Yu, Wan-Qing Chen, Zhi-Ming Shao, Paul E Goss, (2000), “Breast cancer in China,” The Lancet Oncology, Vol.15, June 2014, pp. 280
Expertise summary
Dr Ken Adams from Londonderry Northern Ireland worked in industry for many years as a construction engineer with British Telecom (BT), extending and commissioning telephone exchanges and liaising with contractors, whilst there he completed a degree in mathematics at the University of Ulster (UU). Later, still with BT, he ran the local trade union office for the then National Communications Union.
Ken then obtained a teacher training qualification (PGCE) at Greenwich University. His teaching experience started in 1994 at Westminster College London where he prepared students for university, teaching a variety of courses including access to sports science, Advanced Level Statistics and Advanced Level Mechanics. Having students from diverse ethnic backgrounds and cultures in the same class was both a challenge and a delight.
Upon returning home Ken lectured at the local college including business mathematics at pre-university level. He lectured in Engineering Mathematics for the first and second years of the Electronics and Computing degrees at Magee Campus University of Ulster. He also became a dyslexic tutor for the university helping university students with learning difficulties.
In 2007 he obtained a Ph.D. from the University of Ulster (Topic: Multiple-Valued Logic). The thesis earned Ken the faculty prize as the best Ph.D. of the year. Six international publications flowed from the doctorate.
However he always had the ambition to work abroad and taught mathematics on the NCUK International Foundation Year in Hanoi Vietnam. He currently lectures at the Sino-British College Shanghai. Having taught international students after they arrived in the UK Ken thinks it is a privilege to prepare students for the experience whilst they are still in their home country. Ken very much enjoys the cross cultural adventure that working with Chinese students offers in a dynamic place such as Shanghai.
Ken’s current research interests include data mining and the challenge of big data sets that now exist because of massive information now captured my computers such as by the human genome project. In particular he is fascinated by how many diverse fields such as aggregation, information theory, game theory and artificial intelligence and how they are interacting together to make joint contributions to this new field.
Selected Publication
Adams, Ken (March 1993), "How polite is x?", The Mathematical Gazette 77 (478): 79–80.
Adams K, McGregor RJ, (May 2004). "On the Optimisation of Reed- Muller Expressions," Proceedings of 34th IEEE International Symposium on Multiple-Valued Logic, Toronto, Canada 19-22 May, 2004, 168-176.
Yanushkevich SN, Shmerko VP, Lu DC, Adams K, McGregor RJ, (October 2003), “Arithmetic Spectrum Computing via Taylor Expansion,” The 4th Workshop on Computational Intelligence and Information Technologies, Nis, Serbia, Електронски Факултет, ISBN 86-80135-78-X, Pages 143-146
Adams, K, McGregor RJ, (May 2003), "New Information on the Effectiveness of Different Reed-Muller Algebras on the Representation of Quaternary Functions," Proceedings of, 33rd IEEE International Symp. on Multiple-Valued Logic, Tokyo, May 16-19 2003, 33-39
Adams K, Bell D, Maguire LP, McGregor RJ, (Apr. 2003), "Knowledge Discovery from Decision Tables by the Use of Multiple-Valued Logic," Artificial Intelligence Review , Vol. 19, No. 2, Kluwer, Pages 153-176
Adams K, McGregor RJ, (May 2002). "Comparison of Different Features of Quaternary Reed-Muller Canonical Forms and Some New Statistical Results," Proceedings of 32nd IEEE International Symp. on Multiple-Valued Logic, Boston, USA, May 15-18 2002, 83-88.
Adams, K., Campbell, J.G., Maguire, L.P., Webb, J.A.C., (May 1999), “State Assignment Techniques in Multiple-Valued Logic,“ Proceedings of 29th International Symposium on Multiple-Valued Logic, Freiburg Germany, 20-22 May, 220-225.
Adams, K., “Simple Heuristics for the Choquet Integral Classifier”,Proc. 2nd International Conf. on Electrical, Electronics, Computer Engineering and their Applications (EECE2015) Manila ,Philippines, 2015.
