Statistical Learning

Statistical learning theory understands learning as a function estimation problem on empirical data. The goal are generalizing modells which are able to make predictions for new and unseen data points. Conditions on those generalizations are analyzed, e.g. a finite VC-dimension. This finally leads to risk bounds for classification algoithms. A major idea coming from statistical learning theory is the idea of structural risk minimization. This idea motivates new classification algorithms like Support Vector Machines / Relevance Vector Machines which provide better risk bounds.


RapidMiner (YALE)


Lee, Sangkyun
Molina, Alejandro
Morik, Katharina

Past Master Thesis


Mierswa/Morik/2008a Mierswa, Ingo and Morik, Katharina. About the Non-Convex Optimization Problem Induced by Non-positive Semidefinite Kernel Learning. In Advances in Data Analysis and Classification, Vol. 2, No. 3, pages 241--258, 2008.
Mierswa/2007a Mierswa, Ingo. Controlling Overfitting with Multi-Objective Support Vector Machines. In Proc. of the Genetic and Evolutionary Computation Conference (GECCO 2007; best paper award), 2007.
Mierswa/2007c Mierswa, Ingo. Regularization through Multi-Objective Optimization. In Klinkenberg, Ralf and Mierswa, Ingo and Hinneburg, Alexander and Posch, Stefan and Neumann, Steffen (editors), Proc. of LWA 2007 - Lernen - Wissensentdeckung - Adaptivität, 2007.
Mierswa/2006a Mierswa, Ingo. Evolutionary Learning with Kernels: A Generic Solution for Large Margin Problems. In Proc. of the Genetic and Evolutionary Computation Conference (GECCO 2006), 2006.
Mierswa/2006b Mierswa, Ingo. Making Indefinite Kernel Learning Practical. Collaborative Research Center 475, University of Dortmund, 2006.
Morik/Rueping/2006a Morik, Katharina and Rüping, Stefan. Klassifikations-/Clustermethoden und Konjunkturanalyse. Duncker & Humbolt, 2006.
Morik/Rueping/2006b Morik, Katharina and Rüping, Stefan. An Inductive Logic Programming Approach to the Classification of Phases in Business Cycles. In Heilemann, U. and Weihs, Claus (editors), Classification and Clustering in Business Cycle Analysis, Duncker & Humbolt, 2006.
Scholz/2006a Scholz, Martin. Boosting in PN Spaces. In Johannes Furnkranz, Tobias Scheffer, Myra Spiliopoulou (editors), Proceedings of the 17th European Conference on Machine Learning (ECML-06), Vol. 4212, pages 377--388, Berlin, Germany, Springer, 2006.
Rueping/Scheffer/2005a Ruping, Stefan and Scheffer, Tobias (editors). Proceedings of the ICML 2005 Workshop on Learning with Multiple Views. 2005.
Rueping/2004a Rüping, Stefan. A Simple Method for Estimating Conditional Probabilities in SVMs. In Abecker, A. and Bickel, S. and Brefeld, U. and Drost, I. and Henze,N. and Herden, O. and Minor, M. and Scheffer, T. and Stojanovic, L. and Weibelzahl, S. (editors), Lernen - Wissensentdeckung - Adaptivität, Berlin, Humboldt-Universität, 2004.
Rueping/2004b Rüping, Stefan. Probabilistic SVMs - How Much Scaling Do We Need?. 2004.
Rueping/Morik/2003a Rüping, Stefan and Morik, Katharina. Support Vector Machines and Learning about Time. In IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP'03), 2003.
Rueping/Morik/2003b Rüping, Stefan and Morik, Katharina. Support Vector Machines and Learning about Time. No. 4, SFB475, Universitat Dortmund, Dortmund, Germany, 2003.
Morik/etal/2002a Morik, Katharina and Joachims, T. and Imhoff, M. and Brockhausen, P. and Rüping, S.. Integrating Kernel Methods into a Knowledge-based Approach to Evidence-based Medicine. In Schmitt, Manfred and Teodorescu, Horia-Nicolai and Jain, Ashlesha and Jain, Ajita and Jain, Sandhya and Jain, Lakhmi C. (editors), Studies in Fuzzi- ness and Soft Computing., Vol. 96, pages 71--99, Physica-Verlag, 2002.
Morik/Rueping/2002a Morik, Katharina and Rüping, Stefan. A Multistrategy Approach to the Classification of Phases in Business Cycles. In Elomaa, Taprio and Mannila, Heikki and Toivonen, Hannu (editors), Machine Learning: ECML 2002, Vol. 2430, pages 307--318, Berlin, Springer, 2002.
Rueping/2002a Stefan Rüping. Support Vector Machines in Relational Databases. In Seong-Whan Lee and Alessandro Verri (editors), Pattern Recognition with Support Vector Machines --- First International Workshop, SVM 2002, pages 310--320, Springer, 2002.
Rueping/2002b Rüping, Stefan. Efficient Kernel Calculation for Multirelational Data. In Kokai, Gabriella and Zeidler, Jens (editors), Proceedings der FGML 2002, pages 121-126, Learning Lab Lower Saxony, Hannover, Germany, 2002.
Rueping/2002c Rüping, Stefan. Incremental Learning with Support Vector Machines. No. 18, SFB475, Universitat Dortmund, Dortmund, Germany, 2002.
Rueping/2001a Rüping, Stefan. SVM Kernels for Time Series Analysis. In Klinkenberg, Ralf and Ruping, Stefan and Fick, Andreas and Henze, Nicola and Herzog, Christian and Molitor, Ralf and Schroder, Olaf (editors), LLWA 01 - Tagungsband der GI-Workshop-Woche Lernen - Lehren - Wissen - Adaptivitat, pages 43-50, Dortmund, Germany, 2001.
Rueping/2001b Rüping, Stefan. Incremental Learning with Support Vector Machines. In Cercone, Nick and Lin, T.Y. and Wu, Xindong (editors), Proceedings of the 2001 IEEE International Conference on Data Mining (ICDM '01), pages 641--642, IEEE, 2001.
Morik/etal/99a Morik, Katharina and Brockhausen, Peter and Joachims, Thorsten. Combining statistical learning with a knowledge-based approach -- A case study in intensive care monitoring. In ICML '99: Proceedings of the Sixteenth International Conference on Machine Learning, pages 268--277, San Francisco, CA, USA, Morgan Kaufmann Publishers Inc., 1999.
Rueping/99a Rüping, Stefan. Zeitreihenprognose fur Warenwirtschaftssysteme unter Berucksichtigung asymmetrischer Kostenfunktionen. Universitat Dortmund, 1999.
Schewe/97b Schewe, Sandra. Automatische Kategorisierung von Volltexten unter Anwendung von NLP-Techniken. Fachbereich Informatik, Universitat Dortmund, 1997.
Wermeckes/96a Wermeckes, Thorsten. Erweiterung von COBWEB/CLASSIST zur Behandlung von zeitveranderlichen Daten. Fachbereich Informatik, Universitat Dortmund, 1996.
Robers/95a Ursula Robers. Entwicklung eines wissensbasierten Assistentensystems zur Analyse von Fall--Kontroll--Studien. Universitat Dortmund, Fachbereich Informatik, Lehrstuhl VIII, 1995.
Wessel/95a Stefanie Wessel. Lernen qualitativer Merkmale aus numerischen Robotersensordaten. Fachbereich Informatik, Universitat Dortmund, 1995.