Liddy, two CNLP researchers win best paper award at annual Search Engine Meeting

Wednesday, May 16, 2007, By News Staff

Liddy, two CNLP researchers win best paper award at annual Search Engine MeetingMay 16, 2007Margaret Costello Spillettmcostell@syr.edu

Elizabeth Liddy, Trustee Professor in the School of Information Studies and director of the Center for Natural Language Processing (CNLP) at Syracuse University, along with CNLP software engineers Ozguar Yilmazel and Grant Ingersoll, won the Everett Brenner Prize for the Best Paper at the 2007 Search Engine Meeting, held April 23-24 in Boston. They received the award for their paper “Pro-Active Question Answering.” The prize included a commemorative cup and a monetary award.

The paper examines the practice of question answering, in which a search engine provides not simply a listing of URLs that match keywords input by the user and links to pages that contain undifferentiated types of information, but rather short answers or snippets that address the topics and relationships relevant to the question that was asked. Liddy, along with researchers at CNLP, has extended the practice of question answering to include an inverse function that tracks the questions a user has asked and provides newly acquired information based on what his search history suggests will be useful.

The annual Search Engine Meeting provides a forum and point of reference for those interested in the intricacies of search and retrieval. The meeting draws those with a professional interest in search engines — such as search engine designers and developers — and those interested in applying search engines in their own professional environments. Search is at the heart of information retrieval, and the Search Engine Meeting provides an annual point of reference as to what is happening in this fast-moving and exciting field.

The research was funded by Syracuse Research Corp., an independent, not-for-profit research and development organization dedicated to problems of national significance.

The full text of the paper can be read online at http://www.infonortics.com/searchengines/…liddy.pdf.