Professional Documents
Culture Documents
The Department of Computer Science at the State Uni- Reasoning About the Temporal Structure
versity of New York (SUNY) at Buffalo, one of the old- of Narrative Texts
est computer science departments in the United States to A major part of what it means to understand a story is
grant a Ph.D, has been actively engaged in AI research to know how the different events and states described in
since its establishment in 1965. With half of our full-time the text relate to one another temporally. Although much
faculty members (Shoshana L. Hardt, William J. Rapa- of this information is provided simply by the order of pre-
port, Stuart C. Shapiro, Sargur N. Srihari, and Deborah sentation within the text, a number of other factors play
Walters), more than 40 graduate students, and several as- significant roles in the production of an adequate tempo-
sociated faculty members involved in research programs in ral model. Among these factors are changes in tense, the
the various branches of AI, the department is engaged in a use of the perfect tense, the distinction between progres-
variety of applied and theoretical AI research efforts. The sive and nonprogressive, time adverbials, world knowledge,
different research groups use a variety of AI methodolo- and the inherent aspectual properties of different classes of
gies and tools in a variety of different fields, including nat- predicates. We also believe it is necessary to use some rep-
ural language understanding and computational linguis- resentation of the present moment within the story, that
tics, expert systems, computer vision and pattern recog- is; a narrative “now point.” This narrative now point is
nition, knowledge representation, reasoning, and cognitive moved forward in time as the story progresses in time, and
science. The departmental laboratories include VAXes, it functions as the temporal deictic center to which all the
LISP machines, a SPERRY 7000/40, SUN workstations, a events and situations that occur in the story are related.
Grinnell GMR 274 graphics system, an Imaging Technolo- The major focus of our research is on how the various fac-
gies IP512 image-processing system, and an Adage 3010 tors mentioned here interact with and affect this narrative
graphics system.
Abstract
This article contains reports from the various research groups
Natural-Language Understanding in the SUNY Buffalo Department of Computer Science, Vision
And Computational Linguistics Group, and Graduate Group in Cognitive Science It is or-
ganized by the different research topics. However, it should
be noted that the individual projects might also be organized
Our natural-language understanding projects are focused around the methodologies and tools used in the research, and,
around representing and reasoning about spatial and tem- of course, many of the projects fall under more than one cate-
poral information and belief and knowledge reports, as well gory.
as a knowledge-engineering approach to natural language.
Selection and Use of Image Features Although most of the projects discussed so far have knowl-
in Early Visual Processing edge representation as a major component, the projects
mentioned in this section are directly concerned with it.
One of the important issues in vision is determining which
features of an image should be processed in the early, par-
allel stages of visual processing. Another way of saying this Representation of Visual Knowledge
is that we need to determine what types of image represen- We are investigating methods of representing visual knowl-
tations are most useful for preattentive processing. This edge in a knowledge representation system integrated with