My research interests lies in natural language processing, information
retrieval and machine learning.
Projects
Minspeak Visual Language System for Chinese
(Jan.2006 - present)
Design Chinese iconic Minspeak system. (more to come...)
Speech Act Classification on Emails
(May.2005 - Jul.2005)
Classify emails into different speech act categories (such as Constative,
Commissive, Directive and Acknowledgement) based on different features
extracted by Lemur system and trained by SVM methods. It's the foundational
step towards detecting people's roles in different communications.
Grad. Machine Learning Course Project
(Jan.2005 - Apr.2005)
Implemented a phrase based joint probability model on Chinese-English
translation (referred to A Phrase-Based, Joint Probability Model
for Statistical Machine Translation by Marcu and Wong [pdf])