Brendan O'Connor

I'm interested in machine learning and natural language processing especially when informed by or applied to the social sciences: psychology, economics, political science, etc.

Currently, I'm a computer science grad student at Carnegie Mellon University (CMU), in Noah's ARK group, and an incoming PhD student somewhere next year (though haven't decided where yet).

Contact info.

• Email is brenocon@gmail.com.
• Twitter: @brendan642 (active)
• Blog: anyall.org/blog (moderately active)
• Office: GHC 5414
• Other stuff: facebook/brenocon, linkedin, delicious, aim: brendan642, gtalk: brenocon

Fun blog posts include:

• Comparison of data analysis packages: R, Matlab, SciPy, Excel, SAS, SPSS, Stata
• Statistics vs. Machine Learning, fight!

Previously, I worked with crowdsourced annotations at Dolores Labs / CrowdFlower, and before that, I was at Powerset (now part of Bing), building a "semantic" search engine. ^{[1] [2]}

I was an undergrad at Stanford with a BS/MS in Symbolic Systems: cognitive science, broadly construed. (bsthesis, msthesis)

Recent Research

• From Tweets to Polls: Linking Text Sentiment to Public Opinion Time Series.
  Brendan O'Connor, Ramnath Balasubramanyan, Bryan R. Routledge, and Noah A. Smith.
  Forthcoming at ICWSM-2010.
  (pdf)
• TweetMotif: Exploratory Search and Topic Summarization for Twitter.
  Brendan O'Connor, Michel Krieger, and David Ahn.
  Forthcoming, demo track at ICWSM-2010.
  (pdf)
• Superficial Data Analysis: Exploring Millions of Social Stereotypes.
  Brendan O'Connor and Lukas Biewald.
  In Beautiful Data, ed. Toby Segaran and Jeff Hammerbacher. O'Reilly Media. 2009.
  (blog post) (pdf)
• Cheap and Fast — But is it Good? Evaluating Non-Expert Annotations for Natural Language Tasks.
  Rion Snow, Brendan O’Connor, Daniel Jurafsky, and Andrew Y. Ng.
  Presented at EMNLP-2008.
  (blog post) (pdf) (slides)
  • Using Mechanical Turk for natural language processing (linguistic annotations).

See also a CV / Resume.

Demos

• TweetMotif for summarizing twitter topics. description.
• ARKref, a coreference resolution system.
• KwicTwit - colorized, keyword-aligned kwic-style layout for twitter search.
• Treeviz - quick and dirty parse tree/dependency visualization via graphviz.
• PalinSpeak - ELIZA meets n-grams meets tf/idf. For politics.
• GhettoSearch - a shell script web search engine. Explanation and reddit discussion. The full site is down at the moment.
• read a fortune/quote (code: fun fact, you can write cgi's in shell)

Code