Shan He
Graduate Student, Computer Science
University of Rochester
Research Interests:
I am interested in machine learning techniques in general,
especially those applied to supervised and unsupervised
learning. My thesis work focuses on the statistical
approaches for semantic labeling. I also worked on commonsense and affect reasoning
for visual agents. My advisor is Prof. Daniel
Gildea. I am also organizing the Machine Learning
Seminar for Spring 2006.
Short Bio:
Before coming to Rochester, I got my BS and ME degrees from
Shanghai
Jiao Tong Univerisity in Shanghai, China. I was working in Speech
Recognition group in Department of Electrical
Engineering, under the supervision of Professor Jie Zhu.
CV (pdf).
Contact Information:
Email: shanhe@cs rochester edu
Office: 724 Computer Studies Building
Phone: 585-275-5605
Publication:
- Shan He, Daniel Gildea, "Integrating Cluster Information for
Cross-Frame Semantic Role Labeling", TR 892 ,
Department of Computer Science, University of Rochester,
February 2006.
- Shan He, Daniel Gildea, "Self-training and Co-training for
Semantic Role Labelling", TR 891 ,
Department of Computer Science, University of Rochester,
February 2006.
- Shan He, Daniel Gildea, "Semantic Labeling by Maximum Entropy
Model " TR847, Department of Computer Science, University of Rochester, September 2004.
- Shan He, Dana Ballard, Daniel Glidea, "Building and Tagging with
an Affect Lexicon", TR833, Department of Computer Science, University
of Rochester, March 2004.
- Shan He, Jie Zhu , "A Bootstrap Method for Chinese New Words
Extraction", in Proceedings of ICASSP-2001, Volumn 1, SPEECH-L12: ACOUSTIC & LEXICAL MODELING,
I-581, May 7-11, 2001, Salt Lake City, Utah.
- Jie Zhu, Shan He, "A Study on Construction of Mandarin Voice
Database", in Proceedings of 1999 Japan-China Symposium on Advanced
Information Technology, p55-64, Oct.5-7, 1999, Tokyo, Japan.
- HE Shan, ZHU Jie, "An Iterative Method for Extracting Chinese
Unknown Words", The Chinese Journal of Electronic, Vol.4, 2001.
Author: Shan He
Last modified: Feb. 1, 2006