Accession Number:

ADA361641

Title:

Extensible Multi-Agent System for Heterogeneous Database Association Rule Mining and Unification

Descriptive Note:

Master's thesis

Corporate Author:

AIR FORCE INST OF TECH WRIGHT-PATTERSONAFB OH SCHOOL OF ENGINEERING

Personal Author(s):

Report Date:

1999-03-01

Pagination or Media Count:

107.0

Abstract:

The Probabilities, Expert System, Knowledge, and Inference PESKI System uses the data mining of association rules to fill incompleteness in the knowledge of an expert system. The rules are mined from transactional data sources and then incorporated using an existing link to PESKI. One method of providing a system that will easily allow new data sources to be added is using an information gathering agent-based system. This research first develops a methodology for designing and creating a multi-agent system. It then applies this methodology to design a platform independent means of data mining data sources of any format. The use of agents allows a format specific agent to be used for every data source. The system performs the data mining, then unifies the association rules to present one list of unique results to the parent application.

Subject Categories:

  • Computer Systems
  • Cybernetics

Distribution Statement:

APPROVED FOR PUBLIC RELEASE