AI Seminar
October 28, 2002
Peter Foltz

Applications of Latent Semantic Analysis

Abstract:
Latent Semantic Analysis (LSA) is a computational model of human
knowledge acquisition and representation that provides a method for
extracting the underlying semantic similarity of words and passages
from representative bodies of text. Simulations of psycholinguistic
phenomena show that LSA reflects similarities of human meaning
effectively.  LSA can also be applied to a number of areas in information
retrieval, automated assessment, and natural language processing.  In this
talk, I'll demonstrate a number of applications of LSA including,
information filtering, cross-language information retrieval, automatic
scoring of essay exams, and automated semantic tagging.