SISTA is very excited to announce that on March 3, 2015, the University of Arizona faculty senate voted to approve the formation of a new UA School of Information. This new unit on campus will combine the talent and expertise of faculty and staff in SISTA and the School of Information Resources and Library Science (SIRLS) to create a new, innovative school, with undergraduate and graduate programs that will prepare students to tackle the challenges of a 21st century information based society.
For lots more information, click here!
SISTA Postdoc Peter Jansen has designed and built a science tricorder-like device similar to those used on Star Trek, and has entered this work into “The Hackaday Prize”, a prestigious open source design competition. Out of 800+ competitors, Peter has made it to the final five….and the winner gets a trip to space!
To learn more about his project submission visit – http://hackaday.io/project/1395-open-source-science-tricorder
The competition page is here: http://hackaday.io/prize
Check out the Linguistics Colloquium this Friday for a talk given by Dr. Mihai Surdeanu, Associate Professor in the School of Information: Science, Technology, and Arts
Title: Teaching Computers to Answer Non-Factoid Questions
Date: Friday, September 19, 2014
Time & Location: 3:00 – 4:30 pm in Communication building, Room 311
In this talk, I will describe our work towards teaching computers to answer complex questions, i.e., where the answer is a longer piece of text that explains a complex phenomenon, using linguistic information that is automatically acquired from free text. I will present a robust question answer model for non-factoid questions that integrates multiple sources of information, such as lexical semantics and discourse information, driven by two representations of discourse: a shallow representation centered around discourse markers, and a deep one based on Rhetorical Structure Theory. I will describe how to evaluate the proposed system on two corpora from different genres and domains: one from Yahoo! Answers and one from the biology domain, and two types of non-factoid questions: manner and reason. I will experimentally demonstrate that the discourse structure of non-factoid answers provides information that is complementary to lexical semantic similarity between question and answer, improving performance up to 24% (relative) over a state-of-the-art model that exploits lexical semantic similarity alone. I will further demonstrate excellent domain transfer of discourse information, suggesting these discourse features have general utility to non-factoid question answering.