COMPUTATIONAL INTELLIGENCE RESEARCH GROUP

Within our research group of CTU researchers and students we design and develop new nature inspired algorithms to tackle real world problems mostly in data mining and artificial intelligence. We help companies to solve complex problems such as prediction, classification, recommender systems, or segmentation. Our approach benefits from nature inspired methods and other optimization approaches employed in machine learning to achieve high level of automation and intelligence not only in data mining.
 

RESEARCH AREAS

  • New algorithms, data structures, and infrastructure  for Big data
  • Continuous optimization
  • Data Clustering
  • Data and text mining
  • Genetic algorithms and programming
  • Information visualization
  • Metalearning and ensemble algorithms
  • Predictive modeling
  • Recommender systems
  • Reinforcement learning

 

COOPERATING INSTITUTIONS

 

GRANTS
527/2014 (DF CESNET) Detection of Phishing Attacks in the CESNET Network, main investigator: doc. RNDr. Ing. Marcel Jiřina, Ph.D., 5/2014 - 9/2015.
 

SELECTED PUBLICATIONS
KORDÍK, P. and ČERNÝ, J.: Building Predictive Models in Two Stages with Meta-Learning Templates optimized by Genetic Programming. In:  Proceedings of SSCI 2014, IEEE, 2014.
 
BARTOŇ, T. and KORDÍK, P.: Encoding time series data for better clustering results. In: Int. Joint Conference CISIS'12 - ICEUTE'12 - SOCO'12 Special Sessions, Springer, 2013, Advances in Intelligent Systems and Computing 189, pp. 467-475. ISBN 978-3-642-33017-9.
 
JIŘINA, M.: Utilization of singularity exponent in nearest neighbor based classifier. Journal of Classification, 2013, 30(1), 3-29. ISSN 0176-4268.
 
KORDÍK, P. and ČERNÝ, J.: Self-organization of Supervised Models. In: Meta-Learning in Computational Intelligence, Springer, 2013, Studies in CI 358, pp. 179-223.  ISBN 978-3-642-20979-6.
 
ŘEHOŘEK, T. and KORDÍK, P.: A Soft Computing Approach to Knowledge Flow Synthesis and Optimization. In: Soft Computing Models in Industrial and Environmental Applications, Springer, 2013, pp. 23-32. ISBN 978-3-642-32921-0.
 
BOROVIČKA, T., JIŘINA, M., KORDÍK, P., and JIŘINA, M.: Selecting Representative Data Sets. In: Advances in Data Mining Knowledge Discovery and Applications,  InTech - Open Access Company (InTech Europe), 2012, pp. 43-66. ISBN 978-953-51-0748-4.
 
HÁVA, O., SKRBEK, M., and KORDÍK, P.: Document Classification with Supervised Latent Feature Selection. In: Proceedings of the 2nd International Conference on Web Intelligence, Mining and Semantics, ACM, 2012, pp. 70-74. ISBN 978-1-4503-0915-8.
 
KORDÍK, P. and ČERNÝ, J.: On performance of Meta-learning Templates on Different Datasets. In: The 2012 International Joint Conference on Neural Networks (IJCNN), IEEE, 2012, pp. 1-7. ISBN 978-1-4673-1490-9.
 
KOVÁŘÍK, O. and KORDÍK, P.: Max-min ant system with linear memory complexity. In: Proceedings of the IEEE Congress on Evolutionary Computation (CEC 2012), vol. 1, IEEE, 2012, pp. 1-5. ISBN 978-1-4673-1509-8.
 
KORDÍK, P., KOUTNÍK, J., DRCHAL, J., KOVÁŘÍK, O., ČEPEK, M., and ŠNOREK, M.: Meta-learning approach to neural network optimization. Neural Networks, 2010, 23(4), 568-582. ISSN 0893-6080.
 

CONTACT
Ing. Pavel Kordík, Ph.D.
e-mail: pavel.kordik@fit.cvut.cz

 



Last modified: 21.10.2014, 13:12