About the Knowledge Engineering Group
The Knowledge Engineering Group (KEG) is an informal research group at the University of Economics, Prague. It is centered around the Department of Information and Knowledge Engineering (DIKE) and covers most of its research activities.
This website should give you a brief overview of the research topics addressed, people involved, participation and coordination of projects, regular research seminars and KEG's publications.
Nearest KEG seminar: March 13 2014 (10:30 – 12:00), room 336 RB:
- Recommender Systems for E-commerce and the MODGEN Recommendation Platform: Tomáš Řehořek (FIT ČVUT Praha)
- 2013-11-06 – 2013-11-08: KEG-grown tools for data/text mining and web engineering, such as EasyMiner, GAIN and THD, have been successfully presented within the LinkedTV project exhibition stand at the ICT 2013 conference in Vilnius. See also: ICT 2013 conference
- 2013-10-09: T.Kliegr and M.Dojchinovski co-organized an informal workshop (convened by S.Hellmann from Leipzig, within the LOD2 project) on NIF (NLP Interchange Format) and on benchmarking Wikipedia-Based NER. See also: Workshop webpage
- 2013-10-08: In the News Recommender Challenge (focused on recommending news articles in real-time) organized in conjunction with the ACM Conference on Recommender Systems, the InBeat tool developed by J. Kuchař and T. Kliegr within the LinkedTV project received a runner-up prize: it came second out of 17 participating systems. During the three weeks of the contest, it handled over 20 million recommendation requests. The tool (together with the GAIN platform) is also demoed at the main ACM RecSys conference. See also: Recommender Challenge page, InBeat tool page, RecSys'13 demo session
- 2013-10-07 – 2013-10-08: The local LOD2 team (V. Svátek, J. Mynarz, and colleagues) hosted the plenary meeting of this EU project. See also: Blogpost on the meeting
- 2013-09-23: The Workshop on Data Mining on Linked Data (DMoLD'13), with Linked Data Mining Challenge, has been held during the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, ECML PKDD 2013, in Prague, with P. Berka and V. Svátek among the co-chairs. The workshop attracted the highest number of pre-registered among all the 12 ECML/PKDD workshops, and ignited interesting discussions. See also: Workshop homepage, Blogpost about DMoLD'13
- 2013-09-23: Of the four new PhD students joining the Department from September 2013, two (both supervised by V. Svátek) will join our teams of EU projects: Marek Dudáš (ontology visualization, incl. application on product ontologies) will work in the LOD2 project, and Jan Techl (benchmarking of NE recognition and linking) will work in the LinkedTV project.
- 2013-09-19 – 2013-09-21: Dr. Fazel Famili from NRC, Ottawa, Canada, visited our group and gave a course on the applied perspective of Knowledge Discovery and Management See also: Fazel's homepage, Course page
- 2013-09-12 – 2013-09-13: The EU LOD2 project successfully underwent its (third) annual review.
- 2013-09-06: Jindrich Mynarz and Vojtech Svatek showed as panelists at the (local) conference 'Semantic Web in Business'. See also: Conference program
- 2013-08-15: The long-awaited book, Vol.6 of the Umělá inteligence (Artificial Intelligence) series, eventually appeared; it is probably the most comprehensive book ever presenting to the Czech reader the topics of ontological engineering and semantic web (plus some others). Of its 13 chapters, five are co-authored by KEG members or closely collaborating alumni (P.Berka, M.Labský, J.Rauch, V.Svátek, M.Vacura and O.Zamazal). See also: UI6 book homepage at Academia publishing house
Complete list of history is available as well.
Selected research demos
|I:ZI Miner - mining rich associations from data using a web interface (powered by LISp-Miner)|
|LISp-Miner - stand-alone tool for rich association discovery|
|THD - targeted hypernym discovery, using Wikipedia articles in a live setting|
|PatOMat tools - a collection of tools enabling pattern-based ontology transformation (within the OWL language) and related tasks|
Visit us at our office NB 413, find location by Foursquare venue.
Ontoweb cz page provides information about applied, semantic web and web information extraction research in the area of the Czech Republic.
Semantic Web Pipe is runned to read and syndicate Semantic Web–related weblogs.
Advanced FOAF Explorer tool visualises FOAF profiles.