University of Southern California

Story Representation and Management

The Story Representation and Management projects at ICT are aimed at developing technologies and methodologies for supporting the rapid development of U.S. Army training applications involving the collection, analysis, and use of first-person narratives of real-world experiences (stories). Research on this project focuses on the following problems:

  • Automated story capture -- Technologies for automatically recognizing when people are telling stories about their experiences in written text (e.g. Internet weblogs) using statistical natural language processing
  • Story retrieval interfaces -- Technologies for identifying stories from the Internet and other large text repositories that directly address specific training objectives, using semantic and syntactic text processing techniques
  • Story interpretation -- Techniques for integrating automated commonsense inference into the processing of narrative text documents, and methodologies for creating very large scale commonsense knowledge bases
  • Story-based learning environments -- Technologies and methodologies for integrating real-world experiences told as stories into U.S. Army training applications, including interactive comics for collaborative learning and case-based reasoning methods in simulation environments

This project differs from others in the following ways:

  • Specific focus on the discourse genre of nonfiction narratives of people's experiences
  • Commitment to the integration of large-scale knowledge repositories in the interpretation of textual data
  • Focus on natural language descriptions of narrative events instead of hand-authored formal representations

Tags: automated, experience, interpretation, narrative, story

View All Projects »

Goals

  • Develop natural language processing techniques that are specifically suited to handle the complexities of narrative text
  • Develop methods for applying large-scale knowledge resources to the interpretation of narrative text
  • Develop techniques for incorporating information that is inferred from text into a statistical text processing pipeline
  • Develop large case repositories of stories for use in case-based simulation applications
  • Develop effective methods for textual case retrieval and adaptation in support of case-based simulation