Tuesday, May 04, 2010
11:00 AM
CSB 703
Ph.D. Thesis Proposal
Naushad UzZaman
University of Rochester
Interpreting the Temporal Aspects of Language
Understanding temporal information in text is fundamental for deep language understanding, and key to many applications like question answering, information extraction, document summarization or dialog systems. In this thesis proposal, at first we present our work to extract temporal information from raw text, i.e. extracting events, event features, temporal expressions, and identifying temporal relations between the extracted events and temporal expressions. With the state-of-the-art (or equivalent performance in some subtasks) system to do temporal information processing, in this dissertation, we want to extend it to generate the temporal structure of documents, which gives the representation of events in the timeline and also orders the events in terms of temporal ordering between events. Then we want to use this temporal structure to build an application to help people who have trouble to read and understand. We also want to extend our work to medical domain and use the temporal structure to help caretakers get an overview of patient's medical history.