Knowledge Engineering Report: An Expert System for Selecting Reliability Index

Abstract

This report documents the knowledge encoded in Reliability Index Knowledge Base (RIKB). The knowledge is presented in terms of the conceptualizations of the judgmental knowledge used in various level of decision -making during a consultation. We used a declarative knowledge representation mechanism to encode the knowledge necessary for selecting reliability index. In declarative representation of knowledge, the basic constructs of a knowledge base are production rules and attribute-value pairs. In RIKB, attributes represent properties, and characteristics of reliability indexes that affect the decision -making in selecting reliability indexes for a given Criterion- Referenced Test. The value specifies the specific nature of the attributes in a particular situation (decision-making point). For example, INTENDED-USE (of test score) is an attribute, and the value could be decision, description, or program-evaluation. In the following description, the attributes will be indicated by upper case, and the value will be in lower case. An attribute value can either derived from rules, or directly get from user input. The form rule-i, where i is a number, indicates which rule is used to determine the attribute value. The form question-i, where i is also a number, indicates which question will be asked to get the information for the attribute. Refer to the knowledge base list for the exact wording of the rules, and questions.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Apr 01, 1988
Accession Number
ADA232821

Entities

People

  • Zhongmin Li

Organizations

  • University of California, Los Angeles

Tags

Communities of Interest

  • Human Systems

DTIC Thesaurus Topics

  • Artificial Intelligence
  • California
  • Classification
  • Cognitive Science
  • Contracts
  • Education
  • Engineering
  • Expert Systems
  • Measurement
  • Military Research
  • Psychology
  • Reliability
  • Standards
  • Statistics
  • Students
  • Test And Evaluation
  • Universities

Readers

  • Artificial Intelligence
  • Regression Analysis.