Language Technology Seminar Series

Department of Computer Science and Software Engineering
The University of Melbourne


Title: Intelligent Dictionary Lookup for Japanese

Speaker: Lars Yencken (University of Melbourne)

Location: ICT Building, Room 2.06

Date: 7 July 2006

Time: 12-1pm **Note Unusual Time**

Abstract:

Dictionary lookup for unknown words is particularly slow and cumbersome in orthographic languages such as Chinese or Japanese. First-language speakers of Chinese circumvent this via graphical input methods, but such methods are not available to Japanese, and are still hard for second-language learners. Better lookup methods are evolving, which exploit partial knowledge and error models to convert the problem of lookup to one of search. We present an experiment run to obtain human similarity judgements, and the outcomes and models generated from that experiment. Further, we discuss extension to dictionary search at the word level through the FOKS system, and at the inividual kanji level through the Kansuke kanji lookup system.
Disclaimer: This page, its contents and style, are the responsibility of the author and do not necessarily represent the views, policies or opinions of The University of Melbourne.