1 / 16

Dependency Parsing as a Classification Problem

Learn about linking adjacent words in a classification-based dependency parsing approach. Explore decision lists for German and accuracy levels up to 95% for most languages. Discover challenges with the greedy bottom-up parsing model.

cmodlin
Download Presentation

Dependency Parsing as a Classification Problem

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Dependency Parsing as a Classification Problem Deniz Yuret Koç University İstanbul

  2. Linking Adjacent Words kick the red ball ? ? ?

  3. Linking Adjacent Words kick the red ball

  4. Linking Adjacent Words kick the red ball ? ?

  5. Linking Adjacent Words kick the red ball ? ? ?

  6. Linking Adjacent Words kick the red ball

  7. Linking Adjacent Words kick the red ball ? ? ? ?

  8. Linking Adjacent Words kick the red ball ? ? ?

  9. Linking Adjacent Words kick the red ball ? ?

  10. Linking Adjacent Words kick the red ball

  11. Percentage of adj words linked

  12. Sample decision list for German • If XL1:postag=APPR Then NONE • If X:postag=ART, Y:postag=NN Then L:NK • If X:postag=APPR Then R:NK • If TRUE Then NONE

  13. Sample decision list for German • If XL1:postag=APPR Then NONE • If X:postag=ART, Y:postag=NN Then L:NK • If X:postag=APPR Then R:NK • If TRUE Then NONE APPR-ART-NN

  14. Attribute Selection

  15. Accuracy on adj word links

  16. Contributions • Learning adjacent word dependencies using decision lists and GPA. • Accuracy on adjacent word dependencies between 85%-95% for most languages. • Greedy bottom-up parsing model did not work very well.

More Related