Statistical mining of biological data
The project team develops machine-learning methods for data mining, information visualisation and statistical modelling. Here, machine learning means flexible models that can be used in several application areas.
The methods are developed in the scope of bioinformatics and information-retrieval projects, where we cooperate with projects groups working in the application field. The applications act as test cases for the new methods, and the methods reciprocate by solving problems in the application field.
The research of the group currently focuses on discriminative generative modelling, data fusion by modelling dependencies between data sets, supervised unsupervised learning, and models for defining and extracting 'relevant' signals from data.
The project work is divided between the Department of Computer Science at the University of Helsinki and the Laboratory of Computer and Information Science at Helsinki University of Technology.
Contact person : Professor Samuel Kaski
Homepage : http://www.cis.hut.fi/projects/mi/
Selected publications
A. Klami, S. Kaski. Local Dependent Components. In Zoubin Ghahramani (Ed.), Proceedings of the 24th International Conference on Machine Learning (ICML 2007), pp. 425-433. Omni Press, 2007.
M. Oja, J. Peltonen, J. Blomberg, S. Kaski. Methods for estimating human endogenous retrovirus activities from EST databases. BMC Bioinformatics, 8(Suppl 2):S11, 2007.
J. Venna, S. Kaski. Comparison of visualization methods for an atlas of gene expression data sets. Information Visualization, 6:139-154, 2007.