Induction of Multilingual Morphology with Minimal Supervision

by Assistant Professor of Computer Science Richard Wicentowski "For a majority of the world's languages, resources such as bilingual dictionaries, syntactic parsers, and morphological analyzers are sparse or non-existent because they are expensive to manually create, both in time and money," Wicentowski says. "Unfortunately, complex tools such as voice-recognition and machine-assisted language translation rely heavily on the existence of such resources. This talk will focus on my work in developing automatic methods for acquiring morphological analyzers directly from data with minimal human supervision."