Prejsť na obsah
Oblasti výskumu

The main directions of our research include predictive modeling, cluster analysis, anomaly detection and resembling tasks, whereby we focus on intelligent adaptive approach. The methods range from statistics to machine learning methods, for some kind of problems we have also used biologically inspired computing. We work with static as well as with stream data, in both cases our datasets meet the requirements of Big Data. Currently we evaluate our results in the domain of energy and on datasets generated in bioinformatics research.

Data Streams and Batch Data, Data Clustering, Anomaly Detection, Prediction, Big Data, Specific kinds of data, e.g. those generated in bioinformatics research

Viera Rozinajová
Associate Professor
Mária Lucká
Associate Professor
Covers by her research interests data science in broader sense, in particular she concentrates on advanced methods of predictive modeling, cluster analysis, anomaly detection and optimization. Focuses in her research on efficient algorithms and processing of big data sets, with applications in bioinformatics and energy data. Intelligent data analysis methods include clustering of big data, parallel methods and high performance computing.

Anna Bou Ezzeddine
Mária Bieliková
Explores bio-inspired optimization methods, intelligent data processing techniques with a focus on stream data, self-adapting methods. Aims her research to human interactions on the Web with special emphasis on user modelling and personalization, context awareness, collaborations and usability. This includes research of methods for automated analysis and modelling user feedback, and its evaluation by (multi/group) user studies employing eye trackers.

Gabriela Grmanová
Peter Lacko
Her research is oriented on the field of Data mining. She explores mainly advanced methods of clustering and predictive modeling. His research interests include artificial intelligence, neural networks and parallel and distributed computing.

Jakub Ševcech
Ivan Srba
Focuses in his research on time series data analysis, specifically on representation and feature extraction from time series data for various tasks of data analysis such as classification, anomaly detection or forecasting. Covers by his research interests the area of web-based systems which utilize concepts of collaboration and collective intelligence, in particular he focuses on knowledge sharing (mainly in Community Question Answering systems) and computer-supported collaborative learning.

Selected recent publications

    1. GRMANOVÁ, Gabriela - LAURINEC, Peter - ROZINAJOVÁ, Viera - BOU EZZEDDINE, Anna - LUCKÁ, Mária - LACKO, Peter - VRABLECOVÁ, Petra - NÁVRAT, Pavol
      Incremental Ensemble Learning for Electricity Load Forecasting.
      Acta Polytechnica Hungarica. Vol. 13, No. 2 (2016), s. 97-117.

    1. NEMEC, Radoslav - ROZINAJOVÁ, Viera – LÓDERER, Marek
      Prediction of Power Load Demand Using Modified Dynamic Weighted Majority Method.
      Proceedings of the International Conference on Systems Science 2016 (ICSS 2016), Series Title: Advances in Intelligent Systems and Computing, Vol. 539, Springer, 2017, pp. 36-49, ISBN 978-3-319-48943-8

    1. BOU EZZEDDINE, Anna - LÓDERER, Marek - LAURINEC, Peter - VRABLECOVÁ, Petra - ROZINAJOVÁ, Viera - LUCKÁ, Mária - LACKO, Peter - GRMANOVÁ, Gabriela
      Using Biologically Inspired Computing to Effectively Improve Prediction Models.
      International Journal of Hybrid Intelligent Systems. Vol. 13, no. 2 (2016), pp. 99-112. ISSN 1448-5869.

    1. GRMANOVÁ, Gabriela - ROZINAJOVÁ, Viera - BOU EZZEDDINE, Anna - LUCKÁ, Mária - LACKO, Peter - LÓDERER, Marek - VRABLECOVÁ, Petra - LAURINEC, Peter
      Application of Biologically Inspired Methods to Improve Adaptive Ensemble Learning.
      NaBIC 2015. Advances in nature and Biologically Inspired Computing, Proceedings of the 7th World Congress on Nature and Biologically Inspired Computing, in Pietermaritzburg, South Africa, December 01-03, 2015. 1. vyd. [Cham] : Springer, 2016, pp. 235-246. ISBN 978-3-319-27400-3.

    1. FARKAŠ, Tomáš - KUBÁN, Peter - LUCKÁ, Mária
      Effective Parallel Multicore-optimized K-mers Counting Algorithm.
      SOFSEM 2016: Theory and Practice of Computer Science: 42nd International Conference on Current Trends in Theory and Practice of Computer Science, Harachov, Czech Republic, January 23-28, 2016, Proceedings. 1. vyd. Berlin : Springer, 2016, pp. 469-477. ISBN 978-3-662-49192-8.

    1. ŠEVCECH, Jakub – BIELIKOVÁ, Mária
      Repeating Patterns as Symbols for Long Time Series Representation.
      Journal of Systems and Software. In Press.

  1. SRBA, Ivan - BIELIKOVÁ, Mária
    Why is Stack Overflow Failing? Preserving Sustainability in Community Question Answering.
    IEEE Software. Vol. 33, no. 4 (2016), pp. 80-89.

Important recent research results and research projects

  1. International Centre of Excellence for Research of Intelligent and Secure Technologies and Systems, ITMS: 26240120039 (2014 - 2015)
  2. H2020 NEWTON - Networked Labs for Training in Sciences and Technologies for Information and Communication, (2015 - 2018)
  3. Intelligent analysis of huge datasets using semantic-oriented and bio-inspired methods in parallel environment, (2014 - 2017)

Industry collaboration

  • ATOS Research and Innovation, Madrid, Spain (Tomas Pariente Lobo)
  • Predictive modeling of power load demands using real datasets from Slovakia (common project with ATOS IT solutions and services, Ltd., Sfera, Inc.)

Academy collaboration

  • Faculty of Electrical Engineering and Informatics, Technical University Košice (prof. Ján Paralič)
  • Fraunhofer Institute for Industrial Mathematics, Kaiserslautern, Germany (Dr. Valeria Bartsch)
  • University of Minho, Portugal (Prof. Isabel Ramos)
  • University of J.J. Strossmayer, Osijek, Croatia (Prof. Snjezana Rimac-Drlje)
  • City University Dublin (Dr. Gabriel-Miro Muntean)
  • National Technical University of Ukraine, Applied Mathematics (Assoc.Prof. Yevgeniya Sulema)