Framework for user preference learning

Alan Eckhardt, Peter Vojtáš


PrefWork is a highly customizable framework for user preference model learning. It can use various sources of data - database, csv files, ... It is also easily extendible with new methods, ways of testing the methods, interpreting the results etc. It is still restricted to command line use, though a user interface would be a nice extension.

Where to get it

PrefWork is open-source and is therefore freely accessible. Users are welcome with their feedback.


Contact email:



Research group at the department:

Web Semantization Research Group

Supporting research projects and grants:

AV CR 1ET100300517, MSMT MSM0021620838


  • Eckhardt A.: PrefWork - a framework for the user preference learning methods testing, in ITAT 2009 Information Technologies - Applications and Theory, Kráľova studňa, Slovakia, PONT Slovakia, ISBN: 978-80-970179-2-7, pp. 7-13, September 2009
  • Eckhardt A., Vojtáš P.: Combining various methods of automated user decision and preferences modelling, in MDAI 2009 - The 6th International Conference on Modeling Decisions for Artificial Intelligence, Springer, ISBN: 978-3-642-04819-7, pp. 172-181, 2009
  • Eckhardt A., Vojtáš P.: Evaluating natural user preferences for selective retrieval, in WI 2009 - IEEE/WIC/ACM International Conference on Web Intelligence (WI'09), IEEE Computer Society, ISBN: 978-0-7695-3801-3, pp. 104-107, September 2009
  • Eckhardt A., Vojtáš P.: How to learn fuzzy user preferences with variable objectives, in 2009 IFSA World Congress, ISBN: 978-989-95079-6-8, pp. 938-943, July 2009
  • Eckhardt A., Vojtáš P.: Considering data-mining techniques in user preference learning, in 2008 IEEE/WIC/ACM International Conference on Web Intelligence And Intelligent Agent Technology, Sydney, Australia, IEEE, ISBN: 978-0-7695-3496-1, pp. 33-36, 2008
The content of this web site is licensed under Creative Commons Attribution-NonCommercial 3.0 Czech Republic