| Full text | |
| Author(s): |
Total Authors: 3
|
| Affiliation: | [1] Univ Sao Paulo, Lab Computat Intelligence LABIC, Sao Carlos, SP - Brazil
[2] Western Parana State Univ UNIOESTE, Lab Bioinformat LABI, Foz Do Iguacu - Brazil
[3] Fed Univ Sao Paulo UNIFESP, STI, Sao Jose Dos Campos - Brazil
Total Affiliations: 3
|
| Document type: | Journal article |
| Source: | APPLIED ARTIFICIAL INTELLIGENCE; v. 31, n. 9-10, p. 764-791, 2017. |
| Web of Science Citations: | 0 |
| Abstract | |
Feature selection, an important combinatorial optimization problem in data mining, aims to find a reduced subset of features of high quality in a dataset. Different categories of importance measures can be used to estimate the quality of a feature subset. Since each measure provides a distinct perspective of data and of which are their important features, in this article we investigate the simultaneous optimization of importance measures from different categories using multi-objective genetic algorithms grounded in the Pareto theory. An extensive experimental evaluation of the proposed method is presented, including an analysis of the performance of predictive models built using the selected subsets of features. The results show the competitiveness of the method in comparison with six feature selection algorithms. As an additional contribution, we conducted a pioneer, rigorous, and replicable systematic review on related work. As a result, a summary of 93 related papers strengthens features of our method. (AU) | |
| FAPESP's process: | 09/12963-2 - Application of Multiobjective Genetic Algorithms in the Feature Selection Problem |
| Grantee: | Newton Spolaôr |
| Support Opportunities: | Scholarships in Brazil - Master |
| FAPESP's process: | 12/22608-8 - Use of data complexity measures in the support of supervised machine learning |
| Grantee: | Ana Carolina Lorena |
| Support Opportunities: | Research Grants - Young Investigators Grants |