Complex Systems Computation Research Group - CoSCo
The CoSCo research group studies computational problems in complex systems, especially regarding statistical prediction and modelling. Its research areas include Bayes networks and related graphical probability models, information-theoretical approaches to inference (MDL), analysis of causality, and visualization methods for multi-dimensional data.
The work in the group has a strong emphasis on basic research and it is carried out at the intersection of computer science, information theory and mathematical statistics. In addition, the research works towards a strong application component. The theoretical results on methods are being applied in many fields, such as social sciences, criminology, ecology, medicine, historical research, and industrial applications.
Recent research has focused on analysis of period handwriting, sensory data analysis, visualization of software to select candidates in an election, next-generation search technology, and location-aware services. The members of the group have a wide range of abilities, from theoretical research to excellent programming skills.
To name one concrete example of the group’s broad field of expertise, we can mention the unique B-Course data analysis server (b-course.hiit.fi) developed and maintained by the group. It applies the latest research results from the field of probability modelling. The server has thousands of users world-wide, and the results from the analysis service have been used e.g. for the development of a vaccine against HIV, analysing birdsong, and studying gene data. The group also maintains the portal www.mdl-research.org, which is geared towards gathering the pivotal results of the research into the Minimum Description Length (MDL) theory developed by Jorma Rissanen.
Contact person: Professor Petri Myllymäki.
Homepage: http://cosco.hiit.fi/
Projects
MDL-teoriaan perustuvat kuvasignaalien kohinanpoistomenetelmät (KUKOT)
Probabilistiset menetelmät mikrosirudata-analyysissä (PMMA)
SensorPlanet
Superpeer Semantic Search Engine (Alvis)
Search-Ina-Box (SIB)
Cognitive-Level Annotation using Latent Statistical Structure (CLASS)
Cognitively Inspired Visual Interfaces for Representing Multidimensional
Information (CIVI)
Selected publications
T. Silander, P. Kontkanen and P. Myllymäki: On Sensitivity of
the MAP Bayesian Network Structure to the Equivalent Sample Size Parameter. Pp. 360-367 in the Proceedings of the The 23rd Conference on Uncertainty in Artificial Intelligence (UAI-2007), edited by R. Parr and L. van der Gaag. AUAI Press, 2007.
P. Kontkanen and P. Myllymäki: A linear-time algorithm for computing the multinomial stochastic complexity. Information Processing Letters 103 (2007) 6 (September), 227-233.
P. Kontkanen and P. Myllymäki: MDL Histogram Density Estimation. In Proceedings of the Eleventh International Conference on Artificial Intelligence and Statistics (AISTATS 2007), Puerto Rico, March 2007.
H. Wettig, P. Kontkanen and P. Myllymäki: Calculating the Normalized Maximum Likelihood Distribution for Bayesian Forests. IADIS International Journal on Computer Science and Information Systems 2 (2007) 2 (October).
V. Tuulos, J. Scheible and H. Nyholm: Combining Web, Mobile Phones and Public Displays in Large-Scale: Manhattan Story Mashup. Pp. 37-54 in Proceedings of the Fifth International Conference on Pervasive Computing, edited by A. LaMarca M. Langheinrich and K.N. Truong. Lecture Notes in Computer Science 4480, Springer 2007.