Browse wiki

From Semantic Portal Wiki

Jump to: navigation, search
Knowledge base debugging using apprenticeship learning techniques
Abstract This paper describes the ODYSSEUS apprenti This paper describes the ODYSSEUS apprenticeship program, which is designed to refine and debug knowledge bases for the HERACLES expert system shell. The input to ODYESSEUS is the observable behavior of a specialist solving a problem. ODYSSEUS analyzes the observable behavior using two underlying domain theories. These are an abstract and explicit strategy theory of the problem solving method, called heuristic classification, and an inductive theory based on past problem solving sessions.Following the line of reasoning of an expert in a knowledge-intensive domain and localizing problems within a knowledge bases are themselves knowledge-intensive tasks. To faciliatate bringing diverse sources of knowledge to bear on these problems, two key components of ODYSSEUS, the global and local learning critics, are organized as HERACLES-based expert systems. ODYSSEUS is being developed in the domain of medical diagnosis, a task domain where apprenticeship learining plays a crucial role in the creation of human experts. ial role in the creation of human experts.
Author David C. Wilkins +
Bibtype techreport  +
Institution Knowledge Systems, AI Laboratory +
Key KSL-86-63  +
Modification dateThis property is a special property in this wiki. 1 May 2009 14:03:16  +
Month October +
Note 15 pages. +
Number KSL-86-63  +
Tag Computer science +
Title Knowledge Base Debugging Using Apprenticeship Learning Techniques  +
Tr id KSL-86-63  +
Year 1986  +
Categories Technical Report, Publication, KSL Technical Report
hide properties that link here 
  No properties link to this page.
 

 

Enter the name of the page to start browsing from.
Views
Personal tools
Semantic Web Community
Tetherless World constellation
maintenance
Toolbox