Machine Transliteration

Dr Sarvnaz Karimi

University of Melbourne

Date and time: 11.30am - 12.30pm, Friday 17th Oct, 2008

Venue: 10.08.03 (Building 10, Level 8, Room 3)

Abstract:

Machine transliteration is the automatic process of transforming a word -- usually proper names and technical terms -- from

a source language to a target language. This is particularly challenging when source and target languages do not share the same script. Most

research in this area is centered around methods derived from machine translation, downgraded from the word or phrase level to the character

or substring level. A brief history of approaches applied for a variety of language-pairs will be introduced in this talk. It then focuses on

algorithms developed for machine transliteration between English and Persian.

About the speaker:

Dr Sarvnaz Karimi is a postdoctoral research fellow at NICTA Victoria research laboratory located at the University of Melbourne. The main focus of

her current research is language processing and text information retrieval for biomedical texts. She completed her PhD in computer science in the field of

machine transliteration at RMIT University.


Organizer: Dr. Falk Scholer