Title :
Cooperating to learn: knowledge discovery through intelligent learning agents
Author :
Viktor, Herna L.
Author_Institution :
Dept. of Inf., Pretoria Univ., South Africa
Abstract :
A cooperative multi-agent learning system consists of two or more learners or learning agents that cooperate rather than compete whilst attempting to complete the task at hand. The learners have the ability to learn together thus utilising one another´s strengths and alleviating individual weaknesses. The paper describes the cooperative inductive learning team (CILT) multi-agent learning system that consists of two or more machine learners which induce rules from training examples. By cooperating, the individual results of the machine learners are improved and a team knowledge-base, that contains the best team results, is created
Keywords :
data mining; learning by example; multi-agent systems; cooperative inductive learning team; cooperative multi-agent learning system; intelligent learning agents; knowledge discovery; team knowledge-base; training examples; Africa; Artificial neural networks; Data mining; Informatics; Intelligent agent; Learning systems; Machine learning; Peer to peer computing; Supervised learning; Training data;
Conference_Titel :
MultiAgent Systems, 2000. Proceedings. Fourth International Conference on
Conference_Location :
Boston, MA
Print_ISBN :
0-7695-0625-9
DOI :
10.1109/ICMAS.2000.858521
