Paper: Adaptive String Distance Measures for Bilingual Dialect Lexicon Induction

ACL ID P07-3010
Title Adaptive String Distance Measures for Bilingual Dialect Lexicon Induction
Venue Annual Meeting of the Association of Computational Linguistics
Session Main Conference
Year 2007
Authors

This paper compares different measures of graphemic similarity applied to the task of bilingual lexicon induction between a Swiss German dialect and Standard Ger- man. The measures have been adapted to this particular language pair by training stochastic transducers with the Expectation- Maximisation algorithm or by using hand- made transduction rules. These adaptive metricsshowupto11%F-measureimprove- ment over a static metric like Levenshtein distance.