tiwoli App for iPhone!
Expert for Natural Language Processing & Text Mining heureCLÉA (2012–2016)
- collaboration between narratologists and computer scientists
- goal: automation and analysis of temporal phenomena in literary texts
GlAnCoRe – Global Anaphora and Coreference resolution
Bachelor thesis, March-June 2011.
The thesis and the corresponding source code of my system are publicly available. Unforuntately, the data set which was used for my experiments (the CoNLL 2011 data set) can not be provided as LDC owns the copyright. If you want to use, improve or fork this project, feel free to contact me.
|supervisor||Prof. Dr. Anette Frank|
|topic||Implementation and evaluation of GlAnCoRe, an entity-based coreference resolution system, which combines discourse new detection with coreference resolution to first extract entities from a text and then assign all mentions to corresponding entities based on a machine learning classifier. Both components are optimized separately using feature selection, before they are combined with a meta classifier to counterbalance errors of discourse new detection.|
|resources||source code (bitbucket) | thesis (pdf) | abstract | project page|
Technical Translator for the KDE desktop environment.
- technical translator from 2005 to 2008
- translator for the documentation of various KDE applications (e.g., KNetAttach, KDE user guide)
- German localization team