Welcome!
Our comps group studied methods for "Statistical Machine Translation," or SMT. As you might expect, our goal is to translate between natural languages, like English and Spanish.
Why Machine Translation?
- It's useful. Google translate sees a lot of traffic.
- It's fast and automatic.
- It's cheap.
- It's an interesting problem.
Our Goals:
- Reasonably fast and accurate translation.
- Language agnostic translation (the languages we use shouldn't matter).
Our Sources:
- Knight, Kevin. "A Statistical MT Tutorial Workbook." 1999.
- Brown; Pietra; Pietra; Mercer. "The Mathematics of Statistical Machine Translation: Parameter Estimation." Computational Linguistics. 1993.
- Udupa; Faruquie; Maji. "An algorithmic framework for the decoding problem in statistical machine translation." Computational Linguistics. 2004.
- Niessen; Vogel; Ney; Tillmann. "A DP based Search Algorithm for Statistical Machine Translation." Computational Linguistics. 1998.
- Gawro, Jean Mark. "IBM Model II." San Diego State University.