Hauptnavigation

General

Research

Teaching

Staff

Feature Set Transformations

It is a well known phenomenon that the success of learning depends on the selection input representation. In many cases only a transformation of this representation into another space allows the learning of the inherent patterns. Examples for these transformations are:
  • Feature selection: the selection of a feature subset prevents noisy or redundant features from covering the patterns which should be learned.
  • Feature construction: the construction of new features from given ones allows the explicit representation of feature interactions. Learning methods can make use of this knowledge.
  • Feature extraction: the calculation of features from complex data structures like time series of multidimensional data (e.g. images) is called feature extraction.

Projects

SFB 475 subproject A4
SFB 531 Computational Intelligence

Software

RapidMiner (YALE)

Staff

Klinkenberg, Ralf
Mierswa, Ingo
Pölitz, Christian

Past Master Thesis

Publications

Mierswa/Wurst/2006a Mierswa, Ingo and Wurst, Michael. Information Preserving Multi-Objective Feature Selection for Unsupervised Learning. In Maarten Keijzer and Mike Cattolico and Dirk Arnold and Vladan Babovic and Christian Blum and Peter Bosman and Martin V. Butz and Carlos Coello Coello and Dipankar Dasgupta and Sevan G. Ficici and James Foster and Arturo Hernandez-Aguirre and Greg Hornby and Hod Lipson and Phil McMinn and Jason Moore and Guenther Raidl and Franz Rothlauf and Conor Ryan and Dirk Thierens (editors), GECCO '06: Proceedings of the 8th annual conference on Genetic and evolutionary computation, pages 1545--1552, New York, NY, USA, ACM Press, 2006.
Mierswa/Morik/2005a Mierswa, Ingo and Morik, Katharina. Automatic Feature Extraction for Classifying Audio Data. In Machine Learning Journal, Vol. 58, pages 127--149, 2005.
Mierswa/Morik/2005b Mierswa, Ingo and Morik, Katharina. Method trees: building blocks for self-organizable representations of value series: how to evolve representations for classifying audio data. In Proceedings of the Genetic and Evolutionary Computation Conference GECCO 2005, Workshop on Self-Organization In Representations For Evolutionary Algorithms: Building complexity from simplicity, pages 293--300, New York, NY, USA, ACM, 2005.
Mierswa/Morik/2005c Mierswa, Ingo and Morik, Katharina. Evolutionäre Aufzucht von Methodenbäumen zur Merkmalsextraktion aus Audiodaten. In Informatik Spektrum, Themenheft Musik, Vol. 28, No. 5, pages 381--388, 2005.
Mierswa/Wurst/2005a Mierswa, Ingo and Wurst, Michael. Efficient Case Based Feature Construction for Heterogeneous Learning Tasks. No. CI-194/05, Collaborative Research Center 531, University of Dortmund, 2005.
Mierswa/Wurst/2005b Mierswa, Ingo and Wurst, Michael. Efficient Feature Construction by Meta Learning -- Guiding the Search in Meta Hypothesis Space. In Proc. of the International Conference on Machine Learning, Workshop on Meta Learning, 2005.
Mierswa/Wurst/2005c Mierswa, Ingo and Wurst, Michael. Efficient Case Based Feature Construction for Heterogeneous Learning Tasks. In Alipio Jorge and Luis Torgo and Pavel Brazdil and Rui Camacho and Joao Gama (editors), Proceedings of the European Conference on Machine Learning (ECML 2005), pages 641--648, Berlin, Springer, 2005.
Mierswa/2004a Mierswa, Ingo. Automatisierte Merkmalsextraktion aus Audiodaten. Fachbereich Informatik, Universit\"at Dortmund, 2004.
Mierswa/2004b Mierswa, Ingo. Automatic Feature Extraction from Large Time Series. In Weihs, C. and Gaul, W. (editors), Classification -- the Ubiquitous Challenge, Proc. of the 28. Annual Conference of the GfKl 2004, pages 600--607, Springer, 2004.
Mierswa/2004c Mierswa, Ingo. Automatic Feature Extraction from Large Time Series. 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), Proc. of LWA 2004 - Lernen - Wissensentdeckung - Adaptivitat, 2004.
Mierswa/Morik/2004a Mierswa, Ingo and Morik, Katharina. Learning Feature Extraction for Learning from Audio Data. No. 55/04, Collaborative Research Center 475, University of Dortmund, 2004.
Koepcke/2003a Kopcke, Hanna. Haufigkeitsbasierte Merkmalsgenerierung fur die Wissensentdeckung in Datenbanken. Fachbereich Informatik, Universitat Dortmund, 2003.
Mierswa/2003a Mierswa, Ingo. Beatles vs. Bach: Merkmalsextraktion im Phasenraum von Audiodaten. In LLWA 03 - Tagungsband der GI-Workshop-Woche Lernen - Lehren - Wissen - Adaptivitat, 2003.
Ritthoff/Klinkenberg/2003a Ritthoff, Oliver and Klinkenberg, Ralf. Evolutionary Feature Space Transformation using Type-Restricted Generators. In Cantu-Paz, E. et al. (editors), Proceedings of the Genetic and Evolutionary Computation Conference (GECCO 2003) - Part II, pages 1606--1607, Springer, 2003.
Ritthoff/etal/2002a Ritthoff, Oliver and Klinkenberg, Ralf and Fischer, Simon and Mierswa, Ingo. A Hybrid Approach to Feature Selection and Generation Using an Evolutionary Algorithm. No. CI-127/02, Collaborative Research Center 531, University of Dortmund, Dortmund, Germany, 2002.
Ritthoff/etal/2002b Ritthoff, Oliver and Klinkenberg, Ralf and Fischer, Simon and Mierswa, Ingo. A Hybrid Approach to Feature Selection and Generation Using an Evolutionary Algorithm. In Bullinaria, John A. (editors), Proceedings of the 2002 U.K. Workshop on Computational Intelligence (UKCI-02), pages 147--154, Birmingham, UK, University of Birmingham, 2002.