skip navigation.
- Neural network with back-propagation demo
Created as a demo during a AI course in my Master curriculum, this program
simulates a neural net with back-propagation, allows variable input and shows
an error graph.
- Brill part-of-speech tagger for Dutch
As a final project of a course on corpus linguistics in my Master curriculum,
the rule-based pos-tagger developed by
Eric Brill 
was
slightly changed in order to apply the tagger on Dutch corpora.
- Generalized Suffix Trees
For my Master's thesis, I used generalized suffix trees to find regularities
in corpora of natural language sentences and apply these data structure in the
context of unsupervised grammar induction. This demo allows you to construct
a generalized suffix tree given a set of strings.
- Edit Distance
For my Master's thesis, just as with GSTs I used edit / Levenshtein distance
to find regularities in corpora of natural language sentences. This demo allows
you to calculate and visualize edit distance.
- Alignment-Based Learning
An interesting framework for grammar induction I work on and
with is alignment-based learning. This demo allows
you to induce structure from plain text.
- Cohen's Kappa
Agreement can be measured as percentage of the cases on which the coders or
annotators agree, but it would be desirable to take expected agreement by chance
into account. On this page, Cohen's kappa can be calculated for the case of
more than two annotators (at most 9) with multiple classes.
