David McClosky

I'm now a research scientist at IBM Research working on the DeepQA/Watson project. Previously, I was a postdoc at Stanford NLP working with Chris Manning and Dan Jurafsky. Before that, I was a doctoral student working with Eugene Charniak and Mark Johnson in the Brown Laboratory for Linguistic Information Processing (BLLIP). I'm interested in statistical natural language processing, specifically semi-supervised and unsupervised methods (e.g. self-training) for morphology and syntax induction, and information extraction.

A (most likely outdated) CV.

• April 2013: Many updates to BLLIP Parser including Java and Python SWIG bindings, build improvements, and tokenizer fixes.
• April 26, 2013: Talk at CUNY Natural Language Processing Seminar.
• April 15, 2013: Talk at Columnbia NLP (slides).
• February 2013: The BLLIP Parser is now available under the Apache 2.0 license.
• September 2012: Starting at IBM Research as a Research Staff Member.
• July 2012: Attending ACL 2012 and presenting a paper at EMNLP 2012.
• May 2012: Participated in the SANCL 2012 shared task. Our write-up is here. Consituency parser results are also available on MLComp (the link goes to one of several models we've uploaded).

(older news)

• Email: ude.drofnats.sc@ccmd