Intelligent Systems

1
Intelligent Systems

• Petri Myllymäki

University of Helsinki Department of Computer
Science
2
Mission

• Main objective to develop computationally
  efficient, general-purpose intelligent methods
  for solving large-scale real-world problems
• Basic research focus areas
• information-theoretic modeling
• Minimum Description Length (MDL)
• Normalized Compression Distance (NCD)
• probabilistic graphical models
• Bayesian networks, Causal networks, ICA, Discrete
  PCA,...
• Application-oriented research areas
• signal denoising
• next generation information retrieval methods
• semantic web
• technologies for networked collaborative working
  environments
• neuroinformatics

3
Structure and themes

• Complex Systems Computation Research Group
  (CoSCo)
• Head of the group Professor Petri Myllymäki
  2004?, Professor Henry Tirri ? 2004 (on
  industrial leave as Nokia Research Fellow)
• Senior researchers Jorma Rissanen, Wray Buntine,
  Jaakko Kurhila
• Funding Tekes, Academy of Finland, EU, Industry
• Semantic Computing Research Group (SeCo)
• Head of the group Professor Eero Hyvönen
• Funding Tekes, Industry
• Neuroinformatics Group
• Head of the group Docent Aapo Hyvärinen
• Funding Academy of Finland, HIIT

4
Highlights (scientific, industrial, societal)

• P.Kontkanen, P.Myllymäki, W.Buntine, J.Rissanen,
  H.Tirri, An MDL Framework for Data Clustering. In
  Advances in Minimum Description Length Theory
  and Applications, 2005.
• Research area based on Jorma Rissanens seminal
  work on information-theoretic modelling
• www.mdl-research.org the primary portal to
  MDL-related work
• W. Buntine, J. Löfström, J. Perkiö, S. Perttu,
  V. Poroshin, T. Silander, H. Tirri, A. Tuominen,
  V. Tuulos, A Scalable Topic-Based Open Source
  Search Engine. Web Intelligence, 2004.
• New research area initiated by Wray Buntine,
  focusing on content-based search based on
  statistical text analysis
• www.opensourcesearch.org
• Aino a Finnish search engine (launch in summer
  2007)
• P.Myllymäki, T.Silander, H.Tirri, P.Uronen,
  B-Course A Web-Based Tool for Bayesian and
  Causal Data Analysis. International Journal on
  Artificial Intelligence Tools, 2002.
• b-course.cs.helsinki.fi a publicly available
  data-analysis server with over 15.000 users
  world-wide

5
...More Highlights

• T.Roos, P.Myllymäki, H.Tirri, P.Misikangas,
  J.Sievänen, A Probabilistic Approach to WLAN User
  Location Estimation. International Journal of
  Wireless Information Networks, 2002.
• Probabilistic methods for positioning in wireless
  networks
• Spin-off Ekahau Inc.
• P.Kontkanen, J.Lahtinen, P.Myllymäki, T.Silander,
  and H.Tirri, Supervised Model-Based Visualization
  of High-Dimensional Data. Intelligent Data
  Analysis, 2000.
• Probabilistic multidimensional scaling
• Spin-off BayesIT Inc.
• Election candidate selection machine a public
  service hosted by Helsingin Sanomat, the largest
  newspaper in Finland
• Eero Hyvönen, Eetu Mäkelä, Mirva Salminen, Arttu
  Valo, Kim Viljanen, Samppa Saarela, Miikka
  Junnila, and Suvi Kettula, MuseumFinland
  Finnish Museums on the Semantic Web. Journal of
  Web Semantics, 2005.
• MuseumFinland a public portal to Finnish museums
• Finnish Prime Ministers honourable mention for
  most innovative web application in the Quality
  on the web competition 2004.

6
Future vision

• Probabilistic modelling
• model complexity regularization
• new developments in theory and applications of
  MDL
• theoretical elegance vs. computational efficiency
  
• Information retrieval
• hierarchical models
• cross-lingual information retrieval
• cross-media information retrieval
• Rising focus areas
• sensor fusion
• causal inference
• reinforcement learning, Markov decision processes
• Related research issues concerning enabling
  technologies
• data pre-processing, data visualization, grid
  computing, intelligent web crawling,...