Holger Fröhlich

Background

1996 – 2002: Studies of Computer Science at University of Marburg
1999: Exchange Student at Washington State University / USA
May 2002 – Dec. 2002: Diploma thesis on “Feature Selection for Support Vector Machines by Means of Genetic Algorithms” at the Max-Planck-Institute for Biological Cybernetics, Tübingen, in the group of Prof. Dr. B. Schölkopf
Since March 2003: Research assistent / PhD student at the Centre for Bioinformatics Tübingen (ZBIT) in the group of Prof. Dr. A. Zell
Since June 2006: Postdoctorial researcher at the German Cancer Research Center (DKFZ) in the group of Prof. Dr. A. Poustka (see my new homepage)

Research Interests

Statistical data analysis and Kernel Methods in Bio- and Chemoinformatics, Support Vector Machines

Publications

Journal Articles

 Kernel Functions for Attributed Molecular Graphs - A New Similarity Based Approach To ADME Prediction in Classification and Regression
H. Fröhlich, J. Wegner, F. Sieker, A. Zell, QSAR & Combinatorial Science, 25(4):317 - 326, 2006
Abstract: [pdf]

 Data and Graph Mining in Chemical Space for ADME and Activity Data Sets
J. Wegner, H. Fröhlich, H. Mielenz, A. Zell, QSAR & Combinatorial Science, 25(3):205 - 220, 2006

 Towards Optimal Descriptor Subset Selection with Support Vector Machines in Classification and Regression
H. Fröhlich, J. Wegner, A. Zell, QSAR & Combinatorial Science, 23:311 - 318, 2004
Abstract: [pdf]

 Feature Selection for Descriptor based Classificiation Models: Part I - Theory and GA-SEC Algorithm
J. Wegner, H. Fröhlich, A. Zell, Journal of Chemical Information and Computer Science, 44:921 - 930, 2004

 Feature Selection for Descriptor based Classificiation Models: Part II - Human Intestinal Absorption (HIA)
J. Wegner, H. Fröhlich, A. Zell, Journal of Chemical Information and Computer Science, 44:931 - 939, 2004

 Feature Selection for Support Vector Machines using Genetic Algorithms
H. Fröhlich, O. Chapelle, B. Schölkopf, International Journal on Artificial Intelligence: Special Issue on Selected Papers from the 15th IEEE International Conference on Tools with Artificial Intelligence 2003, 13(4):791 - 800, 2004
[pdf]

Conference Papers (peer reviewed)

Gene Regulatory Network Inference via Regression Based Topological Refinement
Jochen Supper, Holger Fröhlich and Andreas Zell, Asia-Pacific Bioinformatics Conference (APBC 2007), Hong Kong, accepted

Infering Gene Regulatory Networks by Machine Learning Methods
Jochen Supper, Holger Fröhlich, Christian Spieth, Andreas Dräger and Andreas Zell, Asia-Pacific Bioinformatics Conference (APBC 2007), Hong Kong, accepted

Vibration-based Terrain Classification Using Support Vector Machines
C. Weiss, H. Fröhlich and A. Zell, Proc. IEEE/RSJ Int. Conf. Intel. Robots and Systems (IROS 2006), 2006, accepted

Kernel Based Functional Gene Grouping
H. Fröhlich, N. Speer, C. Spieth, and A. Zell, Proc. Int. Joint Conf. on Neural Networks (IJCNN), pp. 6886 - 6891, 2006

Functional Distances for Genes Based on GO Feature Maps and their Application to Clustering
N. Speer, H. Fröhlich, C. Spieth, and A. Zell, Proc. Symp. on Comp. Int. in Bioinformatics and Comp. Biology (CIBCB), pp. 142-149, 2005

 Optimal Assignment Kernels for Attributed Molecular Graphs
H. Fröhlich, J. Wegner, F. Sieker, A. Zell, Proc. Int. Conf. on Machine Learning (ICML), pp. 225 - 232, 2005
Abtract: [pdf]

 Which Features Trigger Action Potentials in Cortical Neurons in Vivo?
H. Fröhlich, B. Naundorf, M. Volgushev, F. Wolf, Proc. Int. Joint Conf. on Neural Networks (IJCNN), pp. 250 - 255, 2005
Abtract: [pdf]

 Functional Grouping of Genes Using Spectral Clustering and Gene Ontology
N. Speer, H. Fröhlich, A. Zell, Proc. Int. Joint Conf. on Neural Networks (IJCNN), pp. 298 - 303, 2005
Abtract: [pdf]

 Assignment Kernels For Chemical Compounds
H. Fröhlich, J. Wegner, A. Zell, Proc. Int. Joint Conf. on Neural Networks (IJCNN), pp. 913 - 918, 2005
Abtract: [pdf]

 Efficient Parameter Selection for Support Vector Machines in Classification and Regression via Model-Based Global Optimization
H. Fröhlich, A. Zell, Proc. Int. Joint Conf. on Neural Networks (IJCNN), pp. 1431 - 1438, 2005
Abtract: [pdf]

 Learning to Detect Proximity to a Gas Source with a Mobile Robot
A. Lilienthal, H. Ulmer, H. Fröhlich, A. Stützle, F. Werner, A. Zell, IEEE/RSJ Int. Conf. on Intelligent Robots and Systems (IROS 2004), pp. 1444 - 1449, 2004

 Feature Subset Selection for Support Vector Machines by Incremental Regularized Risk Minimization
H. Fröhlich, A. Zell, Proc. Int. Joint Conf. on Neural Networks (IJCNN) 2004, 3:2041 - 2046, 2004
Abtract: [pdf]

 Gas Source Declaration with a Mobile Robot
A. Lilienthal, H. Ulmer, H. Fröhlich, A. Stützle, F. Werner, A. Zell, Proceedings of the IEEE International Conference on Robotics and Automation (ICRA 2004), pp. 1430 - 1435, 2004

 Feature Selection for Support Vector Machines by Means of Genetic Algorithms
H. Fröhlich, O. Chapelle, B. Schölkopf, Proc. Int. Conf. on Tools with Artificial Intelligence, pp. 142 - 148, 2003
[pdf]

Thesis

  Feature Selection for Support Vector Machines by Means of Genetic Algorithms, Master's thesis, University of Marburg, 2002 (ps) (PDF-Abstract)

Software

  Feature Selection for SVMs via Genetic Algorithms, RFE, Fisher Score, Relief-F (C++)

  NEW VERSION 3.1: Optimal Assignment Kernels, Optimal Assignment Kernels for reduced graphs, Marginalized Graph Kernels, Hungarian method (JAVA) NEW VERSION (source code)

Address, phone and fax number, email

Eberhard-Karls-Universität Tübingen
Wilhelm-Schickard-Institut für Informatik
Lehrstuhl Rechnerarchitektur
Sand 1
D - 72076 Tübingen
 
Germany


Phone: (+49/0) 7071 / 29 77174 
Fax: (+49/0) 7071 / 29 5091
Email: holger.froehlich at uni-tuebingen.de