This is simply a location for links to things that I happen to come
across which I mention in my notes for students taking the undergrad
NLP course or which I think might be interesting. Notification of
broken links and suggestions for additional links would be welcome.
Links for NLP lectures
Pronoun / anaphora resolution
- Test run directory. Many thanks to Shahzad Khan
for providing these results which are based on the Pang/Lee dataset which in turn comes from (some version of) the IMDb.
Please see Shazad's
README for details and also
poldata.README.2.0. Note that the
tokenised texts are available in the
test directory (renamed from txt_sentoken).
For instance, o0 errors
records the reviews misclassified on the first run.
The first line lists test/pos/cv994_12270.txt: we can see the file itself
News articles etc