Jun-Ping Ng

Researcher + Engineer
Email

email at junping d0t ng

What I Am Doing Now

I am currently working as a software engineer in the Demand Forecasting team at Amazon. It's a fantastic place to be at, and I am glad to have the chance to put into use what I have learnt. Before this, I was in the Machine Learning team at Bloomberg, where I worked on sentiment analysis and document classification.

Research

NLP, IE, QA, Summarization

I am drawn to Natural Language Processing (NLP) and its applications. I am interested in improving the way we access data, and have worked on applications such as question-answering (QA) and multi-document summarization. In my research I seek to build better semantic representations of free text with the aim of improving these NLP applications.

Temporal Processing

As part of work for my doctoral thesis, I worked on identifying and extracting temporal relationships within text. Being able to discern these relationships help improve our understanding of the semantics behind the text. In my thesis, I also leveraged on these relationships to improve text summarization.

Corpus

Text Summarization

I have developed with my colleagues a state-of-the-art multi-document summarization system (SWING). SWING participated in the summarization track of the Text Analysis Conference (TAC) 2011 and was the best performing system measured with the automatic ROUGE metric.

Download

Question-Answering

In earlier work, I have developed an open-source QA system -- QANUS. QANUS is developed to serve as a framework to support rapid prototyping of QA systems, and serve as a foundation on which complete QA systems can be built. Eventually the aim is to grow it into a credible benchmark for QA systems.

Download

Publications

External Bibliography Links

Google scholar
DBLP

Conferences

2015

Better Summarization Evaluation with Word Embeddings For ROUGE, Jun-Ping Ng, Viktoria Abrecht. In Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP), pp 1295-1930.

[pdf] [code]

 

2014

Exploiting Timelines to Enhance Multi-document Summarization, Jun-Ping Ng, Yan Chen, Min-Yen Kan, Zhoujun Li. In Proceedings of the Annual Meeting of the Association for Computational Linguistics (ACL), pp 923--933.

[pdf] [slides] [recording]

 

2013

Exploiting Discourse Analysis for Article-Wide Temporal Classification, Jun-Ping Ng, Min-Yen Kan, Ziheng Lin, Wei Feng, Bin Chen, Jian Su, Chew-Lim Tan. In Proceedings of the Conference on Empirical Methods in Natural Langugage Processing (EMNLP), pp 12-23.

[pdf] [slides]
 

Mining Scientific Terms and their Definitions: A Study of the ACL Anthology, Yiping Jin, Min-Yen Kan, Jun-Ping Ng, Xiangnan He. In Proceedings of the Conference on Empirical Methods in Natural Langugage Processing (EMNLP), pp 780-790.

[pdf]

 

2012

Improved Temporal Relation Classification using Dependency Parses and Selective Crowdsourced Annotations, Jun-Ping Ng, Min-Yen Kan. In Proceedings of the International Conference on Computational Linguistics (COLING), pp 2109-2124.

[pdf] [pdf (a4)] [slides] [data]
 

Exploiting Category-Specific Information for Multi-Document Summarization, Jun-Ping Ng, Praveen Bysani, Ziheng Lin, Min-Yen Kan, Chew-Lim Tan. In Proceedings of the International Conference on Computational Linguistics (COLING), pp 2093-2108.

[pdf] [pdf (a4)] [slides]

 

2011

SWING: Exploiting Category Specific Information for Guided Summarization, Jun-Ping Ng, Praveen Bysani, Ziheng Lin, Min-Yen Kan, Chew-Lim Tan. In Proceedings of the Text Analysis Conference (TAC).

[pdf]

 

2010

Extracting Formulaic and Free Text Clinical Research Articles Metadata using Conditional Random Fields, Sein Lin, Jun-Ping Ng, Shreyasee Pradhan, Jatin Shah, Ricardo Pietrobon, Min-Yen Kan. In Proceedings of the NAACL HLT 2010 Second Louhi Workshop on Text and Data Mining of Health Documents, pp 90-95.

[pdf]

 

Earlier

Dynamic Markov Compression Using a Crossbar-like Tree Initial Structure for Chinese Texts, Ghim-Hwee Ong, Jun-Ping Ng. In Proceedings of the International Conference on Information Technology and Applications (ICITA), pp 407-410.

[pdf]

Exploring the Initial Structures of Dynamic Markov Modeling for Chinese Text Compression, Ghim-Hwee Ong, Jun-Ping Ng. In Proceedings of the International Symposium on Distributed Computing and Applications to Business, Engineering and Sciences, pp 460-463.

Journals

Search Engine Reinforced Semi-supervised Classification and Graph-based Summarization of Microblogs, Yan Chen, Xiaoming Zhang, Zhoujun Li, Jun-Ping Ng. Neurocomputing, Volume 152, pp 274-286.

[details]

Academic Reports

Interpreting Time In Text, Summarizing Text With Time, Jun-Ping Ng. PhD Thesis, National University of Singapore.

[pdf]
 

QANUS: An Open-source Question-Answering Platform, Jun-Ping Ng and Min-Yen Kan. Technical Report

[pdf] [Site]
 

Processing Sentiments and Opinions in Text, Jun-Ping Ng. Survey paper done as part of course work, National University of Singapore.

[pdf]
 

Enhancing Honeypot Stealthiness, Jun-Ping Ng. Master Thesis, National University of Singapore.

[pdf]
 

Dynamic Markov Modeling and Compression for Chinese Textual Data, Jun-Ping Ng. Honours Year Project Report, National University of Singapore.

Posters and Misc.

Interpreting Time In Text, Summarizing Text With Time, Jun-Ping Ng. Research Presentation, Yahoo, New York.

Interpreting Time From Text, Summarizing Text With Time, Jun-Ping Ng. Invited Talk, DSO National Laboratories, Singapore.

 

Towards Timeline Construction: Temporal Relations Between Events, Jun-Ping Ng, Min-Yen Kan, Ziheng Lin. Poster presentation, 6th NExT Workshop, Singapore.

[poster]

Professional

Reviewer

I am currently doing regular reviews for ACM Computing Reviews. I had also served as a reviewer for several conferences.

 

ACM Computing Reviews

Featured reviewer, August 2015.

 

ACL 2016, 2017

EMNLP 2016

NAACL 2016

COLING 2014

IJCNLP 2013

Teaching Assistant

National University of Singapore

2009 to 2013

During my PhD candidature, I taught several courses in areas which I am interested and familiar with. These include Operating Systems, Computer Security, and Artificial Intelligence.

 

Adjunct Lecturer

Nanyang Polytechnic

2009 to 2011

I was an adjunct lecturer at the Nanyang Polytechnic, taking classes on Operating Systems and Computer Security.

Anna University, Chennai.

2011-2013, 1st Class with 78%

It is a long established fact that a reader will be distracted by the readable content of a page when looking at its layout. The point of using Lorem Ipsum is that it has a more-or-less normal distribution of letters.