Synthetic data through combinatorial optimization of pairwise probabilities

The generation of synthetic data is a critical area of research in domains where real data are either not available in large quantities or cannot be directly used. Different techniques have been developed to produce high-quality, realistic synthetic datasets which retain the statistical properties o...

ver descrição completa

Detalhes bibliográficos
Autores: Salvia Hornos, Josep M., Fernàndez Camon, César, Mateu Piñol, Carles
Formato: artículo
Estado:Versión publicada
Fecha de publicación:2026
País:España
Recursos:Varias* (Consorci de Biblioteques Universitáries de Catalunya, Centre de Serveis Científics i Acadèmics de Catalunya)
Repositorio:Recercat. Dipósit de la Recerca de Catalunya
OAI Identifier:oai:dnet:recercat____::932433e207a31ed682424ed2ad9fef2a
Acesso em linha:https://doi.org/10.1007/s41060-026-01063-3
https://hdl.handle.net/10459.1/470089
Access Level:acceso abierto
Palavra-chave:Synthetic data
Data mining
Data exploitation
id ES_80f439db06cebfa8e9736f39ee1a1fb8
oai_identifier_str oai:dnet:recercat____::932433e207a31ed682424ed2ad9fef2a
network_acronym_str ES
network_name_str España
repository_id_str
spelling Synthetic data through combinatorial optimization of pairwise probabilitiesSalvia Hornos, Josep M.Fernàndez Camon, CésarMateu Piñol, CarlesSynthetic dataData miningData exploitationThe generation of synthetic data is a critical area of research in domains where real data are either not available in large quantities or cannot be directly used. Different techniques have been developed to produce high-quality, realistic synthetic datasets which retain the statistical properties of the original data. State-of-the-art results focus on the use of neural networks to capture the latent space extracted from the data. While recent advances in the field of deep learning motivate the application in this new context, this paper proposes a naive baseline to generate synthetic data based on pairwise probabilities. We name the technique DISCO (Discrete Intersection Synthesizer through Combinatorial Optimization), a novel synthetic generator for tabular data. DISCO models data by optimizing the intersection of pairwise probabilities on each generated row in order to resemble the original dataset. Our approach preserves marginal (and pairwise) distributions and as a result, resembles the original data with high fidelity with a very simple approach. Evaluation on various synthetic and real-world datasets as well as regression and classification tasks prove DISCO’s ability to generate high-quality data that rivals state-of-the-art models in both statistical accuracy and machine learning efficacy.Open Access funding provided thanks to the CRUE-CSIC agreement with Springer Nature. Josep Maria Salvia Hornos reports funding from Generalitat de Catalunya AGAUR DI-2024-00020. Cèsar Fernández reports funding by the Spanish MCIN/AEI/10.130- 39/501100011033/, FEDER, UE in project IDs PID2022-138564OAI00 and PID2022-137971OB-I00. Carles Mateu reports that this work was partially funded by the Ministerio de Ciencia e Innovación - Agencia Estatal de Investigación (AEI) (PID2021-123511OB-C31- MCIN/AEI/10.13039/501100011033/ FEDER, UE and RED2022-134219-T).Springer2026info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionhttps://doi.org/10.1007/s41060-026-01063-3https://hdl.handle.net/10459.1/470089https://hdl.handle.net/10459.1/470089reponame:Recercat. Dipósit de la Recerca de Catalunyainstname:Varias* (Consorci de Biblioteques Universitáries de Catalunya, Centre de Serveis Científics i Acadèmics de Catalunya)Inglésinfo:eu-repo/grantAgreement/AEI//PID2022-138564OA-I00info:eu-repo/grantAgreement/AEI/Plan Estatal de Investigación Científica y Técnica y de Innovación 2021-2023/PID2022-137971OB-I00info:eu-repo/grantAgreement/AEI/Plan Estatal de Investigación Científica y Técnica y de Innovación 2021-2023/PID2021-123511OB-C31info:eu-repo/grantAgreement/AEI/Plan Estatal de Investigación Científica y Técnica y de Innovación 2021-2023/RED2022-134219-TReproducció del document publicat a https://doi.org/10.1007/s41060-026-01063-3International Journal of Data Science and Analytics, 2026, vol. 22, 154cc-by (c) Josep Maria Salvia Hornos, Cèsar Fernández Camón, Carles Mateu Piñol, 2026Attribution 4.0 Internationalinfo:eu-repo/semantics/openAccesshttp://creativecommons.org/licenses/by/4.0/oai:dnet:recercat____::932433e207a31ed682424ed2ad9fef2a2026-05-29T05:05:01Z
dc.title.none.fl_str_mv Synthetic data through combinatorial optimization of pairwise probabilities
title Synthetic data through combinatorial optimization of pairwise probabilities
spellingShingle Synthetic data through combinatorial optimization of pairwise probabilities
Salvia Hornos, Josep M.
Synthetic data
Data mining
Data exploitation
title_short Synthetic data through combinatorial optimization of pairwise probabilities
title_full Synthetic data through combinatorial optimization of pairwise probabilities
title_fullStr Synthetic data through combinatorial optimization of pairwise probabilities
title_full_unstemmed Synthetic data through combinatorial optimization of pairwise probabilities
title_sort Synthetic data through combinatorial optimization of pairwise probabilities
dc.creator.none.fl_str_mv Salvia Hornos, Josep M.
Fernàndez Camon, César
Mateu Piñol, Carles
author Salvia Hornos, Josep M.
author_facet Salvia Hornos, Josep M.
Fernàndez Camon, César
Mateu Piñol, Carles
author_role author
author2 Fernàndez Camon, César
Mateu Piñol, Carles
author2_role author
author
dc.subject.none.fl_str_mv Synthetic data
Data mining
Data exploitation
topic Synthetic data
Data mining
Data exploitation
description The generation of synthetic data is a critical area of research in domains where real data are either not available in large quantities or cannot be directly used. Different techniques have been developed to produce high-quality, realistic synthetic datasets which retain the statistical properties of the original data. State-of-the-art results focus on the use of neural networks to capture the latent space extracted from the data. While recent advances in the field of deep learning motivate the application in this new context, this paper proposes a naive baseline to generate synthetic data based on pairwise probabilities. We name the technique DISCO (Discrete Intersection Synthesizer through Combinatorial Optimization), a novel synthetic generator for tabular data. DISCO models data by optimizing the intersection of pairwise probabilities on each generated row in order to resemble the original dataset. Our approach preserves marginal (and pairwise) distributions and as a result, resembles the original data with high fidelity with a very simple approach. Evaluation on various synthetic and real-world datasets as well as regression and classification tasks prove DISCO’s ability to generate high-quality data that rivals state-of-the-art models in both statistical accuracy and machine learning efficacy.
publishDate 2026
dc.date.none.fl_str_mv 2026
dc.type.none.fl_str_mv info:eu-repo/semantics/article
info:eu-repo/semantics/publishedVersion
format article
status_str publishedVersion
dc.identifier.none.fl_str_mv https://doi.org/10.1007/s41060-026-01063-3
https://hdl.handle.net/10459.1/470089
https://hdl.handle.net/10459.1/470089
url https://doi.org/10.1007/s41060-026-01063-3
https://hdl.handle.net/10459.1/470089
dc.language.none.fl_str_mv Inglés
language_invalid_str_mv Inglés
dc.relation.none.fl_str_mv info:eu-repo/grantAgreement/AEI//PID2022-138564OA-I00
info:eu-repo/grantAgreement/AEI/Plan Estatal de Investigación Científica y Técnica y de Innovación 2021-2023/PID2022-137971OB-I00
info:eu-repo/grantAgreement/AEI/Plan Estatal de Investigación Científica y Técnica y de Innovación 2021-2023/PID2021-123511OB-C31
info:eu-repo/grantAgreement/AEI/Plan Estatal de Investigación Científica y Técnica y de Innovación 2021-2023/RED2022-134219-T
Reproducció del document publicat a https://doi.org/10.1007/s41060-026-01063-3
International Journal of Data Science and Analytics, 2026, vol. 22, 154
dc.rights.none.fl_str_mv cc-by (c) Josep Maria Salvia Hornos, Cèsar Fernández Camón, Carles Mateu Piñol, 2026
Attribution 4.0 International
info:eu-repo/semantics/openAccess
http://creativecommons.org/licenses/by/4.0/
rights_invalid_str_mv cc-by (c) Josep Maria Salvia Hornos, Cèsar Fernández Camón, Carles Mateu Piñol, 2026
Attribution 4.0 International
http://creativecommons.org/licenses/by/4.0/
eu_rights_str_mv openAccess
dc.publisher.none.fl_str_mv Springer
publisher.none.fl_str_mv Springer
dc.source.none.fl_str_mv reponame:Recercat. Dipósit de la Recerca de Catalunya
instname:Varias* (Consorci de Biblioteques Universitáries de Catalunya, Centre de Serveis Científics i Acadèmics de Catalunya)
instname_str Varias* (Consorci de Biblioteques Universitáries de Catalunya, Centre de Serveis Científics i Acadèmics de Catalunya)
reponame_str Recercat. Dipósit de la Recerca de Catalunya
collection Recercat. Dipósit de la Recerca de Catalunya
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869411938168995840
score 15,812455

Registos relacionados