Jonathan Elsas

I defended my thesis in 2011 and currently work at Google. Please use the links below to contact me.

Language Technologies Institute, School of Computer Science Carnegie Mellon University.
Professonal: Linkedin, Zerply
Blogs: @jelsas, gplus.to/jelsas, Research Blog, Probably Irrelevant (group IR blog)
FAQ: So you want to study IR?
Code: github.com/jelsas
Data: Ancestry.com Online Forum Test Collection

Contact Information:

email: j...@cs.cmu.edu
office: 6414 GHC
mailing address:

Carnegie Mellon University Language Technologies Institute 5000 Forbes Ave. Gates Hillman Complex 5407 Pittsburgh, PA 15213-3815

Research:

Advisor: Jaime Carbonell
My research areas are:

Information retrieval in complex, structured, social environments. This includes search over large blog corpora, email, or online message boards and forums.
Machine learning for document ranking, particularly fast, scalable algorithms.

Publications:

Jonathan L. Elsas. "The Ancestry.com Online Forum Test Colleciton," CMU Tech Report CMU-LTI-017, 2011. [pdf]
Matthew W. Bilotti, Jonathan L. Elsas, Jaime Carbonell, Eric Nyberg. "Rank Learning for Factoid Question Answering with Linguistic and Semantic Constraints," Proceedings of the 19th ACM International Conference on Information and Knowledge Management (CIKM 2010), 2010.[pdf]
Jonathan L. Elsas, Natalie Glance. "Shopping for Top Forums: Discovering Online Discussion for Product Research," Proceedings of the 2010 Workshop on Social Media Analytics (SOMA 2010), 2010.[pdf]
Jonathan L. Elsas, Susan T. Dumais. "Leveraging Temporal Dynamics of Document Content in Relevance Ranking," Proceedings of the Third ACM International Conference on Web Search and Data Mining (WSDM 2010), 2010. [pdf] [slides pdf] [presentation]
Jonathan L. Elsas. "Search in Conversational Social Media Collections," Proceedings of the Third Annual Workshop on Search in Social Media (SSM 2010), 2010. [pdf]
Jonathan L. Elsas, Pinar Donmez, Jamie Callan, and Jaime Carbonell. "Pairwise Document Classification for Relevance Feedback," Proceedings of the 2009 Text REtrieval Conference (TREC 2009), 2009. [pdf]
Jonathan L. Elsas, Jaime Carbonell. "It Pays to be Picky: An Evaluation of Thread Retrieval in Online Forums," Proceedings of the 32nd Annual International ACM SIGIR Conference on Research & Development on Information Retrieval(SIGIR 2009), 2009. [pdf] [notes on the test collection]
Eytan Adar, Jaime Teevan, Susan Dumais, Jonathan L. Elsas. "The Web Changes Everything: Understanding the Dynamics of Web Content," Second ACM International Conference on Web Search and Data Mining (WSDM 2009), 2009. [pdf], Eytan's presentation, and More at Eytan's site
Best Student Paper
Jaime Arguello, Jonathan L. Elsas, Changkuk Yoo, Jamie Callan, and Jaime Carbonell. "Document and Query Expansion Models for Blog Distillation," Proceedings of the 2008 Text REtrieval Conference (TREC 2008), 2008. [pdf]
Vitor R. Carvalho, Jonathan L. Elsas, William W. Cohen, Jaime G. Carbonell. "Suppressing Outliers in Pairwise Preference Ranking," Proceedings of the 17th annual ACM Conference on Information and Knowledge Management (CIKM 2008), 2008. [pdf]
Vitor R. Carvalho, Jonathan L. Elsas, William W. Cohen, Jaime G. Carbonell. "A Meta-Learning Approach for Robust Rank Learning," Proceedings of the SIGIR 2008 Workshop on Learning to Rank for Information Retrieval (LR4IR 2008), 2008. [pdf]
Jonathan L. Elsas, Jaime Arguello, Jamie Callan, Jaime G. Carbonell. "Retrieval and Feedback Models for Blog Feed Search," Proceedings of the 31st Annual International ACM SIGIR Conference on Research & Development on Information Retrieval (SIGIR 2008), 2008. [pdf] [slides]
Jaime Arguello, Jonathan L. Elsas, Jamie Callan, Jaime G. Carbonell. "Document Representation and Query Expansion Models for Blog Recommendation," Proceedings of the Second International Conference on Weblogs and Social Media (ICWSM 2008), 2008. [pdf]
Best Paper award nominee
Jonathan L. Elsas, Vitor R. Carvalho, Jaime G. Carbonell. "Fast Learning of Document Ranking Functions with the Committee Perceptron," Proceedings of the First ACM International Conference on Web Search and Data Mining (WSDM 2008), 2008. [implementation notes] [pdf] [slides .mov medium] [slides .mov large]
Jonathan L. Elsas, Jaime Arguello, Jamie Callan, Jaime G. Carbonell. "Retrieval and Feedback Models for Blog Distillation," Proceedings of the 2007 Text REtrieval Conference (TREC 2007) [notebook version] [slides]
Best performing group in the TREC 2007 Blog Distillation task
Gary Marchionini, Stephanie W. Haas, Junliang Zhang, Jonathan L. Elsas. "Accessing Government Statistical Information," Computer, vol. 38, no. 12, pp. 52-61, December, 2005. http://doi.ieeecomputersociety.org/10.1109/MC.2005.393
Jonathan L. Elsas. "An Evaluation of Projection Techniques for Document Clustering: Latent Semantic Analysis and Independent Component Analysis," Master's Thesis, UNC-CH School of Information and Library Science, July, 2005. http://hdl.handle.net/1901/208
Efron, M., Elsas, J., Marchionini, G., and Zhang, J. 2004. Machine learning for information architecture in a large governmental website. In Proceedings of the 4th ACM/IEEE-CS Joint Conference on Digital Libraries (Tuscon, AZ, USA, June 07 - 11, 2004). JCDL '04. ACM Press, New York, NY, 151-159. http://doi.acm.org/10.1145/996350.996386
Junliang Zhang, Tim Shearer, Gary Marchionini, Miles Efron, John [Jonathan] Elsas. Relation Browser++: an information exploration and searching tool., DG.O 2004. [pdf]

Other:

Teaching Assistant: 11-741: Information Retrieval, spring 2010; 15-493: Information Retrieval and Web Mining, fall 2007.
Program Committee: SIGIR 2011, SIGIR 2010, NAACL-HLT 2010, SIGIR 2009, ICWSM 2009 Data Challenge Workshop, SSM 2008
Reviewer: WWW-08, TOIS
Co-organizer, Information Retrieval Discussion Series at CMU, 2007-present
Contributor to the Analysis of Social Media Wiki. (dead link -- sorry!)

About me:

I completed my Master's degree in Information Science from the School of Information and Library Science at the University of North Carolina in Chapel Hill. My master's thesis was on the evaluation of two different projection or dimensionality reduction techniques for use in document clustering: Latent Semantic Analysis and Independent Component Analysis. You can see a bit more about what I did while at UNC on my old home page.

Prior to my time at UNC, I worked as a software developer at a large telecommunications company and a large "global services" company on a project enabling VoiceXML services over the telecom network. I was laid-off in 2002 (along with most of the company), found myself back in school in 2003, and realized I really liked it.