UMIACS Computational Linguistics Colloquium, May 9, 2003

Cascaded FST Parsing and Its Support of Information Extraction


Mimi Zhang



UMIACS Computational Linguistics Colloquium

May 9, 2003,
Special time, 11am, special room: CSI 2107

Note: CSI is the new computer science instructional building adjoining AV Williams

This presentation will discuss an approach to cascaded FST parsing and its application in information extraction. An integrated system of shallow and deep parsing, with not only syntactic but also semantic analysis, can provide direct support for advanced information extraction tasks, such as the extraction of both general and specifically defined events from real texts. This type of domain independent parser is proven to be efficient, and can be applicable to developing multiple domain-dependent and/or user-configurable products.


For the colloquium series schedule, see the UMD Computational Linguistics Colloquium Series web page at http://umiacs.umd.edu/~resnik/cl_colloquium/. If you are interested in meeting with the speaker, please contact Philip Resnik (resnik@umiacs.umd.edu).