News
The work I recently completed with collaborators McCallum and Wang on the use of LDA-like models to discover topics and roles in social networks was
presented at the NIPS 2004 Workshop on "Structured Data and Representations in Probabilistic Models for Categorization". A copy of the Technical Report
describing the work can be found here.
I gave a talk at the Center for Natural Language Processing at Syracuse University on November 29, 2004. Here is the talk.
Research
The continuing tidal wave of digital information we are all
creating requires tools that can help us manage it and exploit its
usefulness. My research is in the area of applying statistical
concepts to develop tools that can help us classify, search and
transform human language streams. I've worked in the areas of speech
recognition, speaker recognition, and question answering systems.
I've recently abandoned the Microsoft platform tools for document
creation and reverted to Tex. I knew that I could create standard
documents with Tex but was worried about replacing PowerPoint. The Prosper package has answered my prayers.
Python Bindings for the Lemur Toolkit
The Lemur toolkit for information retrieval
has become much more useful to me now that I developed a set of Python
bindings to it. Besides allowing me to quickly prototype new
information retrieval algorithms using the indexing capabilities of
Lemur, I can also quickly extract information and debug my programs
because of the Python command line interface.
Python bindings for the CMU-Cambridge Statistical Language Modeling Toolkit
I also have Python bindings for the
CMU-Cambridge
Language Modeling Toolkit.
Andrés
Corrada-Emmanuel, Andrew McCallum, Padhraic Smyth, Mark Steyvers, and
Chaitanya Chemudugunta.
Social network analysis and topic discovery for the enron email dataset.
Submitted to the `Workshop on Link Analysis, Counterterrorism and Security' to
take place at the 2005 SIAM International Conference in Data Mining., January
2005.
Andrés
Corrada-Emmanuel and Bruce Croft.
Answer models for question answering passage retrieval.
In Proceedings of the ACM SIGIR 2004, 2004.
Andrew McCallum, Andrés
Corrada-Emmanuel, and Xuerui Wang.
The author-recipient-topic model for topic and role discovery in social
networks: Experiments with enron and academic email.
Technical Report UM-CS-2004-096, Department of Computer Science, UMASS at
Amherst, 2004.
Presented at the NIPS'04 Workshop on `Structured Data and Representations in
Probabilistic Models for Categorization'.
Nasreen AbdulJaleel,
Andrés Corrada-Emmanuel, Qi Li, Xiayong Liu, Courtney Wade, and James
Allan.
UMASS at TREC 2003: HARD and QA.
In Proceedings of the Twelfth Text Retrieval Conference (TREC
2003). NIST, 2003.
Andrés
Corrada-Emmanuel, Bruce Croft, and Vanessa Murdock.
Answer passage retrieval for question answering.
Technical Report IR-283, Center for Intelligent Information Retrieval, 2003.
Frederick Weber,
Barbara Peskin, Michael Newman, Andrés Corrada-Emmanuel, and Lawrence
Gillick.
Speaker recognition on single and multispeaker data.
Digital Signal Processing, 10(1-3):75–92, January 2000.
Andrés
Corrada-Emmanuel, Michael Newman, Barbara Peskin, Lawrence Gillick, and
Robert Roth.
Progress in speaker recognition at Dragon Systems.
In Proceedings of the International Conference on Spoken Language
Processing (ICSLP) 98, November 1998.
Sidney, Australia.
Andrés
Corrada-Emmanuel, Paul Bamberg, Anne Demedts, and Steve Lowe.
Performance of large vocabulary Spanish recognizers in different domains.
Revista de Procesamiento del Lenguaje Natural, 1(19):249–256,
September 1996.
Andrés
Corrada-Emmanuel.
Exact solution for superfluid film vortices on a torus.
Physical Review Letters, 72(5):681–684, 31 January 1994.
Andrés
Corrada-Emmanuel.
Algebraic topology and the quantization of circulation in superfluid helium.
Rapid Communications, Physical Review B, 45(5):2553–2556, 1
February 1992.
Andrés
Corrada-Emmanuel.
Incompressible flows of superfluid films on multiply-connected
surfaces.
PhD thesis, University of Massachusetts at Amherst, 1989.
