|Andreas Rauber - Workshop on Data Analysis|
TEACHING / LEHRE
(link to previous WDA Homepages: WDA 2004 (with photos) in Tatranska Polianka, WDA 2003 (with photos) in Ruzomberok, WDA 2002 (with photos) in Kosice, WDA 2001 in Budapest, WDA 2000 (with photos) in Kosice)
6th Workshop on Data Analysis (WDA2005)June 20-22 2005
SummaryIt seems that after two Austrian-Slovak Student Workshops on Data Analysis (WDA) in Slovakia every third one comes to Hungary. The third WDA took place in Budapest in 2001 and this year the 6th WDA in Abaújszántó, the opening door of the famous Tokaj wine region.
The Workshop, held June 20-22 2005, consisted of four main thematic sessions and traditionally participants presented their contributions and intensively discussed the results achieved with other workshop participants. Moreover, Marek Parali from Technical University of Koaice gave the invited talk speaking about agent-based solution for flexible services, suggesting some possibilities for data analysis techniques to be possibly exploited within this area. The student part of the workshop began with a session on semantic web and ontologies, including deductive and inductive models for fuzzy RDF data. First contribution describes one possible use of ontology extracted from a set of textual documents for improvement of information retrieval. The second paper provides analysis of possibilities for use of DRF in logic programming making use of fuzzy logic, both, from inductive as well as deductive point of view. The third presentation presented the Memborwser, which is a tool for enhancing information access within Weblogs using semantic information extracted from them. The next session started with a contribution that proposed a novel method for visualization of cluster structure and coherent regions of the Self-Organizing Map and continued with paper describing a new feature extraction process for music retrieval. The third session focused on intelligent text processing and text mining methods. One paper dealt with preprocessing of free-text used in medical records in order to transform it into a structured, XML format. The second paper presented a way to integrate the JBowl system supporting text mining and information retrieval (developed at the Technical University of Kosice) with the GridMiner system (developed at the Vienna University of Technology). The last contribution in this session described experiments with new word-sense disambiguation method based on Wordnet lexical database. The workshop ended up with a theoretical session on data and knowledge representation. Finally, we would like to thank Marek Parali for giving us an interesting and stimulating invited talk and also to all participants of the workshop for their contributions and very fruitful discussions after each presentation, making this year workshop a very successful event.
Below some photos, taken during the seminar and the trip to Boldogkoei Var, a wine-tasting, and the Conference Dinner in an old Tokaji winecellar. (Click on the images to enlarge the thumbnails.)
Trip to Boldogkoei Var