I am a computer scientist with Johns Hopkins University
and I have appointments with the
Applied Physics Laboratory (APL) and the
Human Language Technology Center of Excellence (HLTCOE).
My main body of work has been in multilingual text retrieval and information extraction (mining information from unstructured text).
Projects that I have worked on include:
- LITESABER: Text analytics for low-resource languages. I develop entity analysis tools for the DARPA LORELEI program which aims to support situational awareness during crisis response scenarios such as the 2010 Haitian earthquake.
- Apache JOSHUA (incubating): Statistical machine translation in many languages. Matt Post and I have built
language packs for translating foreign text to English in 60+ languages. Java 1.8 is the only software dependency.
- Knowledge Base Population: I helped organize the inaugural NIST Text Analysis Conference Knowledge Base Population track in 2009. My group at the JHU HLTCOE has been active in research in linking named entities to knowledge bases, and we created a publicly releasable test collection for cross-language entity linking of persons names.
- HAIRCUT: The Hopkins Automated Information Retriever for Combing
Unstructured Text. HAIRCUT is a prototype
system that uses a variety of indexing terms (e.g., words, stems, or character n-grams).
I have participated in numerous international evaluations including:
the Text REtrieval Conference
- An Army Research Office
(ARO) project investigating an optimization technique called
automated memoization. This software engineering tool enables
programmers to improve the run-time performance of their code by
storing results of repetitive calculations without changing a
single line of code. Downloads are available for
I am a senior member of the Association for
Computing Machinery (ACM) and a member of the
Association for Computational Linguistics (ACL).
- Paul McNamee, James Mayfield, Samantha Y. Rowe, Alexander K. Rowe, Hannah L. Jackson, and Megan Baker.
High Recall Text Retrieval for Public Health Systematic Review. To appear at FLAIRS-30, 2017.
- Shane Bergsma, Paul McNamee, Mossaab Bagdouri, Clay Fink, and Theresa Wilson.
'Language Identification for Creating Language-Specific Twitter Collections.'
NAACL-HLT 2012 workshop on Language and Social Media (LSM-12), Montreal, Canada, June 2012.
Learn about our language-identification dataset.
- Paul McNamee, James Mayfield, Dawn Lawrie, Douglas W. Oard, and David Doermann,
Cross Language Entity Linking,
Proceedings of the 5th International Joint Conference
on Natural Language Processing (IJCNLP-2011), Chiang Mai, Thailand, November 2011.
Our test set is available here.
- Paul McNamee, Charles Nicholas, and James Mayfield,
Addressing Morphological Variation in Alphabetic Languages,
Proceedings of the 32nd Annual International Conference on Research
and Development in Information Retrieval (SIGIR-2009), Boston, MA,
pp. 75-82, 2009.
- Paul McNamee,
Textual Representations for Corpus-Based Bilingual Retrieval,
PhD Thesis, University of Maryland Baltimore County, December 2008.
- Paul McNamee, Rion Snow, Patrick Schone, and James Mayfield,
Learning Named-Entity Hyponyms for Question Answering.
Proceedings of the 3rd International Joint Conference
on Natural Language Processing (IJCNLP-2008), Hyderabad, India,
pp. 799-804, January 2008.
- James Mayfield, Paul McNamee, Christine Piatko, and Claudia Pearce,
Lattice-based Tagging Using Support Vector Machines.
Proceedings of the Twelfth International ACM Conference on Information and
Knowledge Management (CIKM 2003), pp. 303-308, November 2003.
- Paul McNamee and James Mayfield,
Comparing Cross-Language Query
Expansion Techniques by Degrading Translation Resources. In the
Proceedings of the 25th Annual International Conference on Research
and Development in Information Retrieval (SIGIR-2002), Tampere,
Finland, pp. 159-166, 2002.
More detailed lists of my scholarly publications are available here or from Google Scholar.
I teach as an adjunct faculty member in the
Johns Hopkins University's
Engineering and Applied Science Program for Professionals,
program in computer science.
The program offers students the opportunity to earn a masters
degree by taking courses at several facilities in the Baltimore and Washington D.C. area.
- Current: 605.646: Natural Language Processing. Covers subjects including text classification, language modelling,
sequential tagging, parsing, lexical semantics, relation extraction, machine translation, and automated question
- Current: 605.744:
Information Retrieval covers the automatic indexing and
retrieval of unstructured textual information.
- Former: 605.435:
Distributed Development on the World Wide Web covers web
topics such as HTML, Applets, CGI scripts and servlets, and
605.201: Introduction to Programming Using Java is a
course designed for students without the programming
experience needed to continue studies in the JHU part-time
program. Topics include the basic syntax, control
methods and data structures of the language.
This course is a replacement for
605.201: Introduction to C++ Programming
- Former: 605.451:
Principles of Artificial Intelligence introduces
various methods, languages, and techniques in Artifical
Intelligence including search, natural language, logic, Lisp,
Prolog, and neural networks.
- Former: 605.452:
Artificial Intelligence Programming expands on 605.451 by
covering examples of AI programs in greater detail and having
students work on projects in search, natural language,
simulation, game-playing, all using the Common Lisp language.
I enjoy paddling and playing Euro-style board games (e.g.,
Taluva, Le Havre, or Fire in the Lake). On 10/29/17 I played
Here I Stand, a six-faction game modeling the spread of the Protestant Reformation in Europe 500 years less 2 days ago.
Paul McNamee: http://pmcnamee.net/