|
Title:
|
MULTI AGENT KNOWLEDGE DISCOVERY |
|
Author(s):
|
María C. Gaya , Ignacio Giráldez |
|
ISBN:
|
972-99353-0-0 |
|
Editors:
|
Pedro Isaías and Nitya Karmakar |
|
Year:
|
2004 |
|
Edition:
|
2 |
|
Keywords:
|
Multi agent knowledge discovery; theory synthesis. |
|
Type:
|
Short Paper |
|
First Page:
|
1085 |
|
Last Page:
|
1088 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
A Multi Agent Decision System (MADES) is a Multi Agent System built for decision making, where a single decision is output by the system as a group, although internally many decisions may be made locally by the component agents (Giráldez, 97a). The use of the IAO (Intelligent Agents Organization) model (Giráldez, 97b) for MADES has produced accuracy results that improve the results obtained by monolithic systems (Giráldez, 99a). But the other parameter used in the KDD quality measure, the explanatory power, has not been improved yet. The improvement of the explanatory power would mean that the theory synthesized by the MADES in the KDD task is greater than the explanatory power of the theories that might be synthesized using the knowledge obtained locally from a single data source. The first step needed to achieve this goal is an algorithm for synthesizing a single global theory from knowledge discovered locally by the component agents of a MADES mining distributed data sources. |
|
|
|
|
|
|
