David Martens

Expertise

Dr. ir. Martens is part-time lecturer at the Department of Business Administration and Public Management at Hogeschool Gent, Association Universiteit Gent, Belgium, and post-doctoral researcher at the Department of Decision Sciences and Information Management at Katholieke Universiteit Leuven, Belgium. Since 2008, dr. Martens is also active as data analysis and risk management consultant.

Previous clients include Dexia, AXA, Bank Van Breda and Vanbreda International, where he has done projects on:

  • PD and LGD model development
  • Validation (backtesting and benchmarking) of PD and LGD models
  • Pilar 1 stress testing
  • Business continuity management
  • Fraud detection
Regularly, dr. Martens tutors professionals, as well as master students at K.U.Leuven, on the mentioned topics; and frequently presents at both national (e.g. data mining garden) and international conferences (e.g. Conference on Credit Scoring and Credit Control).

The work of David Martens is recommended by :
Dr. ir. Tony Van Gestel, Basel II director at Dexia:
''I worked with David as a jury member during his PhD thesis and hired him for consultancy services afterwards. David has a broad knowledge and is a fast learner. He works efficient and interacts fluently with people. His work and attitude were much appreciated by the team and me.''

Prof. dr. Bart Baesens, professor at K.U.Leuven:
''I got to know David as his supervisor at K.U.Leuven. David is bright, young, and talented researcher. He also impressed me with his consulting skills and tutoring abilities. I would strongly recommend him either as a research fellow, or consulting partner.''

Academic

In his research he has focused on the development of comprehensible, intuitive classification techniques, with main application Basel II compliant systems. The research has been published in a variety of high quality international journals, as well as in the national press (e.g. De Standaard, De Morgen, Knack, Radio 1).

dr. ir. David Martens holds following degrees:

  • PhD in Applied Economics from K.U.Leuven (2008)
    "Building Acceptable Classification Models for Financial Engineering Applications"
    Finalist of the international KDD dissertation award
  • Master of Business Administration from Reims Management School (2005)
  • Master in Civil Engineering in Computer Science from K.U.Leuven (2003)

Research

Some of his publications are:

  • Comprehensible credit scoring models using rule extraction from support vector machines, Martens D, Baesens B, Van Gestel T, Vanthienen J, 2007, European Journal of Operational Research, vol. 183, no. 3, pp. 1466 - 1476.
  • Forecasting and analyzing insurance companies' ratings, Van Gestel T, Martens D, Feremans D, Baesens B, Huysmans J, Vanthienen J, 2007, International Journal of Forecasting, vol. 23, no. 3, pp. 513 - 529.
  • An Overview and Framework for PD Backtesting and Benchmarking, Castermans G, Martens D, Van Gestel T, Hamers B, Baesens B, 2007, Credit Scoring and CreditControl X, Edinburgh (U.K.), July 30th 2007.
  • Classification with Ant Colony Optimization, Martens D, De Backer M, Haesen R, Vanthienen J, Snoeck M, Baesens B, 2007, IEEE Transactions on Evolutionary Computation, vol. 11, no. 5, pp. 651 - 665.
  • Decompositional Rule Extraction from Support Vector Machines by Active Learning, Martens D, Baesens B, Van Gestel T, 2008, IEEE Transactions on Data and Knowledge Engineering.
A full list of his publications can be found on his academic website.