The course will cover different aspects of corpus analysis regarding the comparison of corpora/subcorpora in the area of register analysis, diachronic analysis, and contrastive analysis (mostly German English). We will have a look at different methods ranging from traditional corpus linguistic analysis to data mining and machine learning techniques, from data exploration to data analysis and interpretation.
- How can we identify and extract relevant features?
- Corpus exploration
- Features selection
- Corpus query
- Feature extraction
- How can we make sense of the extracted features?
- Visualization
- Classification
- Interpretation
The course will provide theoretical insights as well as practical experience.
More information on the course website: http://fedora.clarin-d.uni-saarland.de/unserwiki/doku.php?id=teaching:ss_2015:hs_comparing_corpora
see also: Übung zum Hauptseminar, Dr. Hannah Kermes, |