Speaker: Dan Tappan, Ph.D student in CS department
Time & location: 2/2704 Friday noon in SH124

Title:

Knowledge-Based Spatial Reasoning for Automated Scene Generation from Text
Descriptions

Abstract:

Understanding text is a trivial task for literate humans.  For computers,
however, it is extremely difficult due to (among other things) a lack of
knowledge about language and the world and an intelligent reasoning
mechanism to process such resources.  As a result, computational
approaches to text understanding generally lack common sense and suffer
from poor performance.  This talk presents a system that addresses a
limited subset of the problem.  It extracts the explicit information in
rudimentary text descriptions of static, spatial scenes of a zoo
environment, integrates it with implicit, background information from a
commonsense knowledge base, reasons over the combined representation, and
renders a set of corresponding graphical interpretations.  From these
depictions, it extracts new information that feeds back into the original
description to augment the understanding further.  This multidimensional
test-and-evaluation framework addresses cognitive, linguistic, and
computational issues of knowledge representation and reasoning over space,
which apply to many applications in artificial intelligence.