Version 1
: Received: 7 October 2020 / Approved: 8 October 2020 / Online: 8 October 2020 (09:21:34 CEST)
Version 2
: Received: 23 December 2020 / Approved: 24 December 2020 / Online: 24 December 2020 (08:59:19 CET)
How to cite:
Batista, J.; Cabral, A.; Vasconcelos, M.; Vanneschi, L.; Silva, S. Improving Land Cover Classification Using Genetic Programming for Feature Construction. Preprints2020, 2020100168. https://doi.org/10.20944/preprints202010.0168.v2
Batista, J.; Cabral, A.; Vasconcelos, M.; Vanneschi, L.; Silva, S. Improving Land Cover Classification Using Genetic Programming for Feature Construction. Preprints 2020, 2020100168. https://doi.org/10.20944/preprints202010.0168.v2
Batista, J.; Cabral, A.; Vasconcelos, M.; Vanneschi, L.; Silva, S. Improving Land Cover Classification Using Genetic Programming for Feature Construction. Preprints2020, 2020100168. https://doi.org/10.20944/preprints202010.0168.v2
APA Style
Batista, J., Cabral, A., Vasconcelos, M., Vanneschi, L., & Silva, S. (2020). Improving Land Cover Classification Using Genetic Programming for Feature Construction. Preprints. https://doi.org/10.20944/preprints202010.0168.v2
Chicago/Turabian Style
Batista, J., Leonardo Vanneschi and Sara Silva. 2020 "Improving Land Cover Classification Using Genetic Programming for Feature Construction" Preprints. https://doi.org/10.20944/preprints202010.0168.v2
Abstract
Genetic Programming (GP) is a powerful Machine Learning (ML) algorithm that can produce readable white-box models. Although successfully used for solving an array of problems in different scientific areas, GP is still not well known in Remote Sensing. The M3GP algorithm, a variant of the standard GP algorithm, performs Feature Construction by evolving hyper-features from the original ones. In this work, we use the M3GP algorithm on several sets of satellite images over different countries to create hyper-feature from satellite bands to improve the classification of land cover types. We add the evolved hyper-features to the reference datasets and observe a significant improvement of the performance of three state-of-the-art ML algorithms (Decision Trees, Random Forests and XGBoost) on multiclass classifications and no significant effect on the binary classifications. We show that adding the M3GP hyper-features to the reference datasets brings better results than adding the well-known spectral indices NDVI, NDWI and NBR. We also compare the performance of the M3GP hyper-features in the binary classification problems with those created by other Feature Construction methods like FFX and EFS.
Computer Science and Mathematics, Algebra and Number Theory
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Received:
24 December 2020
Commenter:
João Batista
Commenter's Conflict of Interests:
Author
Comment:
After an initial round of reviews, this manuscript was extended with details such as: - A more detailed explanation of the M3GP algorithm; - Inclusion of more information about the climate in the Study Areas; - Commentaries on the popularity of each original feature in the creation of hyper-features for each problem; - Commentaries on the impact of the hyper-features in each class (rather than just overall accuracy) .
Commenter: João Batista
Commenter's Conflict of Interests: Author
- A more detailed explanation of the M3GP algorithm;
- Inclusion of more information about the climate in the Study Areas;
- Commentaries on the popularity of each original feature in the creation of hyper-features for each problem;
- Commentaries on the impact of the hyper-features in each class (rather than just overall accuracy) .