Evolving decision trees with beam search-based initialization and lexicographic multi-objective evaluation

作     者：Basgalupp, Marcio P. Barros, Rodrigo C. de Carvalho, Andre C. P. L. F. Freitas, Alex A. 

作者机构：Univ Fed Sao Paulo BR-12231280 Sao Jose Dos Campos SP Brazil Univ Sao Paulo BR-13560970 Sao Carlos SP Brazil Univ Kent Canterbury CT2 7NF Kent England 

出 版 物：《INFORMATION SCIENCES》 (信息科学)

年 卷 期：2014年第258卷

页      面：160-181页

核心收录：

学科分类：12[管理学] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 08[工学] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：Fundacao de Amparo a Pesquisa do Estado de Sao Paulo (FAPESP) Conselho Nacional de Desenvolvimento Cientifico e Tecnologico (CNPq) 

主　　题：Decision tree Lexicographic optimization Machine learning Multi-objective evolutionary algorithm 

摘      要：Decision tree induction algorithms represent one of the most popular techniques for dealing with classification problems. However, traditional decision-tree induction algorithms implement a greedy approach for node splitting that is inherently susceptible to local optima convergence. Evolutionary algorithms can avoid the problems associated with a greedy search and have been successfully employed to the induction of decision trees. Previously, we proposed a lexicographic multi-objective genetic algorithm for decision-tree induction, named LEGAL-Tree. In this work, we propose extending this approach substantially, particularly w.r.t. two important evolutionary aspects: the initialization of the population and the fitness function. We carry out a comprehensive set of experiments to validate our extended algorithm. The experimental results suggest that it is able to outperform both traditional algorithms for decision-tree induction and another evolutionary algorithm in a variety of application domains. (C) 2013 Elsevier Inc. All rights reserved.