University of Konstanz
Graduiertenkolleg / PhD Program
Computer and Information Science

Dr. Nicolas Cebron

Doctoral Student in the PhD program from 01.04.2005 to 01.05.2008.

advisors

1. Prof. Dr. Michael Berthold
2. Prof. Dr. Ulrik Brandes

organisational data

Room: Z 809
Tel.: +49 (0)7531 / 88-4719
E-mail: ncebron@gmail.com
Other Resources: Personal webpage
picture

project description

Data Mining in Bioinformatics

The development of high-throughput imaging instruments - e.g. flourescence microscope cameras - resulted in them becoming the major tool to study the effect of agents on different cell types. These devices are able to produce about 55,000 images per day; until recently, visual inspection by a domain expert was the only way to distinguish between 'active' and 'nonactive' cells.

The aim of this project is to design classifiers that are able to learn the differences between cell types. As we are dealing with a large amount of unlabeled data, the expert should label only a small subset to train the classifier. Choosing randomly drawn examples from the dataset would render the classifier biased towards the underlying distribution of the different kinds of cells.

Therefore, we try to apply the concept of 'active learning' to this task, where our learning algorithm has control over which parts of the input domain it receives information about. This concept is very similar to the human form of learning, whereby problem domains are examined in an active manner.

The objective of this thesis is to develop new concepts and algorithms based on the idea of active learning for miscellaneous data-mining algorithms in order to build stable classifiers in the field of bioinformatics.

publications

The following list of publications covers only those, which are or were published during participation at the Graduiertenkolleg / PhD program.

Articles in Journals

20092008
2009
2008
  • Cebron, N., Berthold, M., Active learning for object classification: From exploration to exploitation, Journal of Data Mining and Knowledge Discovery, Vol. 18, No. 2, pp. 283-299, 2008. File on Springer's Website
  • Cebron, N., Berthold, M., Adaptive prototype-based fuzzy classification, Fuzzy Sets and Systems, Vol. 159, No. 21, pp. 2806-2818, 2008.

Conference Papers

2009200720062005
2009
  • Meinl, T., Cebron, N., Gabriel, T., Dill, F., Kötter, T., Ohl, P., Thiel, K., Wiswedel, B., Berthold, M., The Konstanz Information Miner 2.0, Proceedings of OSDM 2009, pp. 26-31, Bangkok, Thailand, 2009.
2007
  • Berthold, M., Cebron, N., Dill, F., Gabriel, T., Kötter, T., Meinl, T., Ohl, P., Sieb, C., Thiel, K., Wiswedel, B., Knime: The Konstanz Information Miner, Studies in Classification, Data Analysis, and Knowledge Organization (GfKL), Freiburg, Germany, 2007, Springer-Verlag. abstract
  • Berthold, M., Cebron, N., Dill, F., Gabriel, T., Kötter, T., Meinl, T., Ohl, P., Sieb, C., Thiel, K., Wiswedel, B., Open Source Data Pipelining für Interaktive Datenexploration, Aktuelle Trends in der Softwareforschung (do it.software-forschungstag 2007), Mathematical Intelligencer, Fellbach, Germany, 2007, MFG Stiftung Baden-Württemberg. abstract
2006
2005

Technical Reports

2007

Phd Theses

2008

curriculum vitae

2004 - 2005 Participation in special admissions procedure for the PhD Program in Computer Science. 
1999 - 2004 Studies of Computer Science at the University of Applied Sciences in Braunschweig/Wolfenbuettel, Germany.
Degree: Diplom-Informatiker (FH).
Internship at British Telecom, Ipswich / UK: Development of an estimation procedure for task time prediction.