Lecture Notes in Computer Science, 2009, Volume 5774/2009, 54-65, DOI: 10.1007/978-3-642-04143-3_6

Multi Criteria Decision Methods for Coordinating Case-Based Agents

Beatriz López, Carles Pous, Pablo Gay and Albert Pla

View Related Documents

Abstract

There is an increasing interest on ensemble learning since it reduces the bias-variance problem of several classifiers. In this paper we approach an ensemble learning method in a multi-agent environment. Particularly, we use genetic algorithms to learnt weights in a boosting scenario where several case-based reasoning agents cooperate. In order to deal with the genetic algorithm results, we propose several multi-criteria decision making methods. We experimentally test the methods proposed in a breast cancer diagnosis database.

Keywords  Ensemble Learning - Case-Based Reasoning - Multi Criteria Decision Making

Fulltext Preview

Image of the first page of the fulltext document