Next: Tagger Evaluation
Up: Background and Test Setups
Previous: Background
We suggest three types of tests:
- :
tests allowing to assess the impact of different statistical tagging
methods on the results, by comparing the performance of different
taggers on the same training and test data, using the same tagset;
- :
tests allowing to assess the impact of tagset modifications on the
results, by using different versions of a given tagset on the same
texts; differences between the versions of the tagset are documented
and classified, and the impact of each modification is tested;
- :
tests allowing to assess the impact of perceived linguistic
differences between training texts and test (or: application) texts on
the results, by using texts from different text types in training and
testing, tagsets and taggers being unchanged otherwise.
The test setups and the objectives underlying each test are described
in more detail in the following.
We suggest three series of tests: