User Modeling and Register Theory: A Congruence of Concerns

Abstract

Sophisticated computer systems using natural language to interact with people are now becoming widespread. These systems need to communicate with an increasingly varied user community, across an ever more extensive range of situations. Just as for human-human interaction, no single style of generated text is adequate across all user types and all situations. Generation systems can only be effective if they appropriately 'taylor' their phrasing, text content, and organization according to the situation and to the abilities and requirements of the intended readers. This paper presents new work in 'tailoring' that addresses the phrasing problem: how to best express the propositional content that has been chosen by a text planner, given a user and situation. Importantly, this paper shows how relevant linguistic studies can be bought to bear the problem of user modeling and tailoring. In particular, we would like to show that the concerns of register theory are very close to some of the concerns of user modeling, and that aspects of the theory can guide us in studies in user modeling. Based on this specific linguistic theory, we propose a methodology to systematically study the problem of tailoring phrasing. Tailoring, User modeling, Register theory, Generation systems.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Nov 01, 1990
Accession Number
ADA269676

Entities

People

  • Cecile L. Paris
  • John A . Bateman

Organizations

  • University of Southern California

Tags

Communities of Interest

  • Biomedical
  • Ground and Sea Platforms

DTIC Thesaurus Topics

  • Communities
  • Computer Science
  • Computers
  • Data Analysis
  • Demographic Cohorts
  • Dictionaries
  • Digital Circuits
  • Errors
  • Expert Systems
  • Grammars
  • Information Science
  • Language
  • Linguistics
  • Natural Languages
  • Physicians
  • Pilot Studies
  • Specifications

Readers

  • Agent-Based Social Robotics and Mobile-Assisted Learning in Virtual Environments.
  • Artificial Intelligence
  • Computational Linguistics