January 2nd, 2007 by scharl
We are looking for a student with sound programming skills, preferably in Python and/or Java, and a background in statistics interested in writing an undergraduate thesis in the area of natural language processing and text mining (to be supervised either at Vienna University of Economics and Business Administration or Graz University of Technology). For generating and structuring its knowledge base, IDIOM requires algorithms to automatically identify and tag relations between information entities. Thus the aims of this thesis are:
- Identification of weaknesses in the current approach based on trigger phrases,
- Improvement of the existing approach by applying a Sequential Probability Model,
- Evaluation of the performance of the refined architecture.
The developed methodology will help to improve crucial parts of IDIOM’s core architecture. Thus, final results will be needed around August 2007. An excellent thesis completed within this timeframe will be rewarded with a bonus between 300 and 1,000 EUR.
Contact: Dr. Albert Weichselbraun
Institut für Informationswirtschaft
Augasse 2-6, A-1090 Wien