University of Cambridge > Talks.cam > NLIP Seminar Series > The Weakest Link: Detecting and Correcting Errors in Learner English

The Weakest Link: Detecting and Correcting Errors in Learner English

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact NLIP Seminars.

There is a pre-theoretic intuition that stretches of text with low probability should correspond to errors of the type made by non-native writers. Unfortunately, perfectly well-formed text can be quite unlikely according to a contiguous n-gram model. I’ll describe how I built a dependency-structure model for English in which unlikelihood is a much better indicator of error and present some preliminary results.

This talk is part of the NLIP Seminar Series series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.

 

© 2006-2019 Talks.cam, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity