Multi-generation multi-criteria feature construction using Genetic Programming
Created by W.Langdon from
gp-bibliography.bib Revision:1.8098
- @Article{MA:2023:swevo,
-
author = "Jianbin Ma and Xiaoying Gao and Ying Li",
-
title = "Multi-generation multi-criteria feature construction
using Genetic Programming",
-
journal = "Swarm and Evolutionary Computation",
-
volume = "78",
-
pages = "101285",
-
year = "2023",
-
ISSN = "2210-6502",
-
DOI = "doi:10.1016/j.swevo.2023.101285",
-
URL = "https://www.sciencedirect.com/science/article/pii/S2210650223000585",
-
keywords = "genetic algorithms, genetic programming, Feature
construction, Overfitting, Multi-generation,
Multi-criteria",
-
abstract = "The purpose of feature construction is to create new
high level features from the original features. When
Genetic Programming (GP) is applied to wrapper-based
feature construction, especially when the samples size
is small, GP generally overfits the training set and
generalizes poorly with the deepening of evolution.
Overfitting has attracted wide attention in some
classification models, however, it is not commonly
studied in the field of feature construction. In this
paper, a Multi-Generation feature construction method
(MG) is developed to preserve the solutions produced by
multiple generations of GP. A Multi-Criteria feature
construction method (MC) is introduced to use a
multi-criteria evaluation function to evaluate GP
individuals. Combining the above two methods, a
Multi-Generation Multi-Criteria feature construction
method (MGMC) is proposed. Experiments on fourteen
datasets show that the proposed MG and MC methods can
improve the classification performance and overcome
overfitting problems of traditional feature
construction methods in most cases. The combined MGMC
method further improves the classification performance
and achieves the best results",
- }
Genetic Programming entries for
Jianbin Ma
Xiaoying (Sharon) Gao
Ying Li
Citations