Christian Monson
Ph.D. Student Language Technologies InstituteSchool of Computer Science
Carnegie Mellon University 4524 Newell Simon Hall
5000 Forbes Ave.
Pittsburgh, PA 15213 (412) 268 - 4757
cmonson@cs.cmu.edu
April 15, 2008 was my Ph.D defense. My thesis describes ParaMor, an algorithm for unsupervised induction of natural language morphology. Pending revisions to the thesis document, my anticipated graduation date is August 2008. With graduation imminent, I am currently looking for a job. I'd love to hear from you about any natural language processing position that might fit my background.
While my thesis was on morphology, I am broadly interested in computational understanding of natural language: an area that for me includes everything from machine translation, natural language understanding, and parsing to information retrieval and speech recognition.
Read on for a description of my thesis work.
Analyzing the internal structure of words is a vital first step in natural language processing tasks from machine translation to speech recognition. But developing systems to analyze the internal, or morphological, structure of words for new languages is challenging and time consuming. In my thesis work I seek to overcome the bottleneck of morphological acquisition.
My algorithm, ParaMor, automatically discovers the morphological structure of a language from nothing more than text in that language. In the spring of 2007 I entered ParaMor in Morpho Challenge, a competition measuring the performance of data-induced morphological analysis systems. In English, ParaMor outperformed a state-of-the-art baseline system, Morfessor, placing 4th overall. In German, a combined ParaMor-Morfessor system placed 1st.
My thesis advisors are Jaime Carbonell, Alon Lavie , and Lori Levin. With Ron Kaplan of PARC and PowerSet rounding out my thesis committee.