Studies of Verbal Problem Solving. 1. Two Performance-Aiding Programs

Abstract

Two computer programs were written to provide on-line aiding to human problem solvers. Both programs were written in time-shared BASIC, and were designed for 'membership' problems. In this kind of problem, these are several English sentences and implicit in the sentences are various relations; the task is to infer a membership structure that is compatible with all the logical constraints. Membership problems may be cast in various settings, such as a murder mystery where a culprit is to be identified. One program (FIRST) was based on Findler's 'Universal Puzzle Solver' concept; the other (GABE) used Wang's theorem-prover logic. In both programs, the human operator converted English problem sentences to logical membership relations. The programs kept track of all relations entered, indicated when more data inputs were needed, and scored whether a correct answer was achieved. Of the two programs, FIRST appears to be most feasible with ordinary college subjects. It accepts logical inputs in a near-English format, and shows current logical status of a problem via tabular arrays of X's and O's. The present version of GABE used a strict 'p, q, r' logical notation; college subjects find this difficult and unsatisfactory.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Sep 01, 1977
Accession Number
ADA047193

Entities

People

  • Joseph W. Rigney
  • Nicholas A. Bond Jr.
  • William Gabrielli

Organizations

  • University of Southern California

Tags

Communities of Interest

  • Biomedical

DTIC Thesaurus Topics

  • Air Force
  • Civilian Personnel
  • Cognition
  • Computer Programs
  • Computers
  • Contracts
  • Education
  • Educational Psychology
  • Information Processing
  • Management Personnel
  • Military Research
  • Notation
  • Personnel Management
  • Psychology
  • Schools
  • Students
  • Training

Readers

  • Artificial Intelligence
  • Computational Linguistics
  • Computer Science.

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference
  • AI & ML - Machine Learning Algorithms
  • AI & ML - Machine Translation