NLP Publications

• Don Blaheta, ``Function tagging'' (PhD thesis), 28 Aug 2003. HTML (abstract and links), PDF (110 pages/790K), PDF for two-sided printing (123 pages/795K). Defence slides: PDF (56 pages/281K).
• Don Blaheta, ``Binominal each'' (A.M. thesis), 1 May 2003. HTML (abstract and links), two-sided PDF (88 pages/410K), one-sided PDF (77 pages/403K).
• Don Blaheta, ``Function tagging'' (thesis proposal), 22 Aug 2002. HTML (abstract and links), PS (57 pages/806K), PDF (57 pages/327K). Talk slides: PDF (36 pages/91K).
• Don Blaheta, ``Handling noisy training and testing data'', Proceedings of the 7th conference on Empirical Methods in Natural Language Processing, Philadelphia, July 2002, pp. 111--116. HTML (abstract and links), PS (6 pages/321K), PDF (6 pages/102K).
• Don Blaheta and Mark Johnson, ``Unsupervised learning of multi-word verbs'', ACL Workshop on Collocation, Toulouse, July 2001, pp. 54--60. HTML (abstract), PS (7 pages/127K), PDF (7 pages/89K).
• Don Blaheta and Eugene Charniak, ``Assigning Function Tags to Parsed Text'', Proceedings of the 1st Annual Meeting of the North American Chapter of the Association for Computational Linguistics, Seattle, May 2000, pp. 234--240. HTML (abstract), PS (7 pages/143K), PDF (7 pages/170K).
• Don Blaheta and Eugene Charniak, ``Automatic Compensation for Parser Figure-of-Merit Flaws'', Proceedings of the 37th Annual Meeting of the Association for Computational Linguistics, College Park, MD, June 1999, pp. 513--518. HTML (abstract), PS (6 pages/195K), PDF (6 pages/154K).

Software

• My feature-tree-based function tagger

Other

• Appendix D of my thesis, entitled "The principle of indifference", was a short history-of-math article on the origins of that usually mis-cited foundational principle of probability theory (aka "principle of insufficient reason"). PDF (5 pages/102K).
• "The Lambda Calculus", a set of notes I took from a lecture on that topic by Shriram Krishnamurthi on 12 Oct 2000. This writeup has been in the first page of Google hits for "lambda calculus" for most of that time. PDF (8 pages/152K).