Jonathan L. Elsas Information Science Master's Student, School of Information and Library Science, University of North Carolina at Chapel Hill jelsas--AT--email.unc.edu Resume: (doc) (pdf)

---

• I have completed the SILS Master's program and have started the PhD program at the Language Technologies Institute in the School of Computer Science at Carnegie Mellon University. Please visit my new home page for more information.
• Publications
• Research Interests
  • Text Mining Toolkit: a Java-based toolkit used for parsing and tokenizing HTML documents, applying unsupervised learning techniques (clustering), and analyzing the resulting clusters. This toolkit is used for my current topic detection work (see below). See the user's manual for a more detailed explanation of the toolkit's capabilities. A new version with enhanced clustering and document processing capabilities is in the works.
  • Automatic topic detection in large corpora, using statistical methods of text processing. This work stems from my involvement in the GovStat project, and can be read about in the paper: Machine learning for information architecture in a large governmental website.
  • Converting unstructured or semi-structured text to machine processable data
  • Mapping from the concept-space to the information-space, and back.
• Older GovStats Work. Some of these demos are outdated, unfinished, and generally goofy. But, feel free to explore.
  • WebBuilder
  • DataHunter
  • DataHunter2
  • HTML Analysis for BLS
  • HTML Analysis for EIA

Last updated: November 17, 2004