Vincent Ng

Some historical information about Cornell CS

Juris Hartmanis and the early days of Cornell CS
Cornell CS faculty circa 1995
Upson's familiar quotations
Women in Computer Science

Welcome to my homepage! As far as I know, there is no other page that is linked to this page. Actually this is no longer true, since Kiri and Lillian made a link to this page from the NLP group homepage! Well, I believe it is still safe for me to assume that you know who I am when you are reading this!

Currently there is nothing in this page. This is partly because I am reluctant to spend my time on designing a nice page for myself, and partly because I'd like you to let me know what you want to know about me and what you want me to put in this page. So don't forget to sign my guestbook before you leave!

I am a member of the Cornell Natural Language Processsing Group. My general research area is machine learning of natural language. In particular, I am interested in

applying machine learning techniques to build NLP systems,

statistical modeling for NLP problems, and

understanding the difficulties of natural language learning.

Current and Past Projects

Information Extraction. I am working on RIPTIDES, a text summarization system supported by information extraction. For more information about the IE project, follow this link.

Relevant publications:

Multi-document Summarization via Information Extraction.
Michael White, Tanya Korelsky, Claire Cardie, Vincent Ng, David Pierce, and Kiri Wagstaff.
First International Conference on Human Language Technology Research (HLT-01), 2001.
Detecting Discrepancies and Improving Intelligibility: Two Preliminary Evaluations of RIPTIDES.
Michael White, Claire Cardie, Vincent Ng, Kiri Wagstaff, and Daryl McCullough.
2001 Document Understanding Conference (DUC-01), 2001.
Detecting Discrepancies in Numerical Estimates Using Multidocument Hypertext Summaries.
Michael White, Claire Cardie, Vincent Ng, and Daryl McCullough.
Second International Conference on Human Language Technology Research (HLT-02), 2002.

Noun Phrase Coreference Resolution. Currently I am extending Kiri's clustering framework for noun phrase coreference to incorporate syntactic constraints and discourse-level preferences. Kiri is examining the effects of constraints on the clustering algorithm.

Relevant publications:

Improving Machine Learning Approaches to Coreference Resolution.
Vincent Ng and Claire Cardie.
40th Annual Meeting of the Association for Computational Linguistics (ACL-02), 2002.
Identifying Anaphoric and Non-Anaphoric Noun Phrases to Improve Coreference Resolution.
Vincent Ng and Claire Cardie.
19th International Conference on Computational Linguistics (COLING-02), 2002.
Combining Sample Selection and Error-Driven Pruning for Machine Learning of Coreference Rules.
Vincent Ng and Claire Cardie.
2002 Conference on Empirical Methods in Natural Language Processing (EMNLP-02), 2002.
Learning Noun Phrase Anaphoricity to Improve Coreference Resolution: Issues in Representation and Optimization.
Vincent Ng.
42nd Annual Meeting of the Association for Computational Linguistics (ACL-04), 2004.

Question Answering. I am investigating the use of statistical and linguistic knowledge in answering general-knowledge, domain-independent natural language questions. The Q&A task that we are looking at is the one defined in the TREC-8 QA Track. Here is a paper that described the work we have done so far.
- Examining the Role of Statistical and Linguistic Knoweledge Sources in a General-Knowledge Question Answering System.
  Claire Cardie, Vincent Ng, David Pierce, and Chris Buckley.
  Sixth Applied Natural Language Processing Conference (ANLP-2000), 2000.

Semi-Supervised Natural Language Learning. I have also spent some time looking at issues in bootstrapping natural language classifiers from small datasets.

Relevant publications:

Weakly Supervised Natural Language Learning Without Redundant Views.
Vincent Ng and Claire Cardie.
Human Language Technology Conference of the North American Chapter of the Association for Computational Linguistics (HLT-NAACL), 2003.
Bootstrapping Coreference Classifiers with Multiple Machine Learning Algorithms.
Vincent Ng and Claire Cardie.
2003 Conference on Empirical Methods in Natural Language Processing (EMNLP-03), 2003.

Here is a recent picture of myself (just in case you really don't know who I am), and if you really want to know more about me ...

visitors since 11/10/00.

Vincent Ng < Email : yung@cs.cornell.edu >

Last modified: Sun Jul 4 23:17:56 2004