Types of Explanations.

Abstract

Many people in AI want to use explanations in their work, but nobody (including us) knows enough about them. Although explanation has been the foundation of much AI research, little attention has been paid to two fundamental questions one would like to be able to answer about explanation: What are the kinds of things that need to be explained and what are the types of explanations? This report is an attempt to map out that territory by developing a categorization scheme that covers a substantial variety of anomalies and explanations. We have collected a large group of real-world explanations and studied their similarities in order to understand what categories they fall into. In this paper the anomalies which need to be explained are divided into three broad classes: intentional, social and material anomalies. We discuss the three classes, and present the different kinds of explanations that can apply to each. Keywords: Explanation, Categorization, Anomaly types, Story understanding.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Mar 01, 1987
Accession Number
ADA183253

Entities

People

  • Alex Kass
  • David Leake

Organizations

  • Yale University

Tags

Communities of Interest

  • Biomedical
  • C4I
  • Human Systems

DTIC Thesaurus Topics

  • Artificial Intelligence
  • California
  • Cognitive Science
  • Commerce
  • Computer Science
  • Education
  • Electrical Engineering
  • Electronic Mail
  • Governments
  • Human Behavior
  • Information Processing
  • Market Economy
  • Materials
  • Military Research
  • New York
  • Personality
  • Systems Management

Readers

  • Systems Analysis and Design
  • Theoretical Analysis.