Yliopiston etusivulle In English
Helsingin yliopisto
clt234: Natural Language Processing Applications - lukuvuosi 2009-2010

Yhteystiedot

Nykykielten laitos

PL 24 (Unioninkatu 40)
00014 HELSINGIN YLIOPISTO

Puhelin +358 (09) 1911 (vaihde)
Faksi +358 (09) 191 28313

6. Chunking and Chunkers 1.

  • Lecture notes
  • Further reading
  • Practical work
    • In IDLE, run this NLTK application:
      >>> nltk.app.chunkparser()
      
    • When the demo starts, the sentence in the lower panel shows "Confidence/NN" in red and "the/DT pound/NN" in red.
    • In the upper panel, you type regular expressions for NP chunks. Start with just:
      {<NN>}
      This makes "Confidence/NN" turn green, but "the/DT pound/NN" is still partly red.
    • Edit the grammar by adding a determiner:
      {<DT><NN>}
      This makes "the/DT pound/NN" turn green, but now "Confidence/NN" is red again.
    • Make the determiner optional:
      {<DT>?<NN>}
      Now both "Confidence/NN" and "the/DT pound/NN" are green but "another sharp dive" is partly red.
    • Challenge: Can you make all the red phrases turn green, by improving the regexp grammar for NP chunks?
© 2008-2010 Graham Wilcock

Hae laitoksen sivuilta:

Laitoksen etusivulle | Tiedekunnan etusivulle | Yliopiston etusivulle

Copyright © 2003-2005 Helsingin yliopisto. Kaikki oikeudet pidätetään.