Datos sintéticos: Introducción a técnicas generativas y evaluación de calidad

Cleves Leguízamo, Diego Andrés

Datos sintéticos: Introducción a técnicas generativas y evaluación de calidad

dc.contributor.advisor	, Javier Mauricio
dc.contributor.author	Cleves Leguízamo, Diego Andrés
dc.contributor.corporatename	Universidad Santo Tomás
dc.contributor.orcid	https://orcid.org/0009-0003-5914-4156
dc.date.accessioned	2026-02-06T11:42:32Z
dc.date.available	2026-02-06T11:42:32Z
dc.date.issued	2026-02-04
dc.description	El presente trabajo aborda el estudio de datos sintéticos desde su concepción teórica y generación. Se propone la implementación de diversos modelos con el fin de sintetizar datos categóricos y cuantitativos, luego se comparan de acuerdo a su capacidad de enmascarar datos (propensión), sus medidas de semejanza estadística y tiempo de ejecución. Los resultados mostraron que simular variables categóricas con base a reglas, que representan sus dependencias en la realidad es el mejor método para simularlas. No obstante, a las variables numéricas no fue posible sintetizarlas de manera adecuada, los modelos propuestos no capturaron la cópula adecuadamente. A manera de conclusión se indica dónde se fallo y las oportunidades de mejora disponibles.
dc.description.abstract	This work studies synthetic data from its theoretical conception to its generation. Several models are implemented to synthesize categorical and numerical data and are compared in terms of data masking capability (propensity), statistical similarity, and execution time. The results indicate that rule-based simulation is the most effective approach for categorical variables, while numerical variables could not be adequately synthesized due to the models’ inability to capture the copula structure. The conclusions discuss the identified limitations and potential improvements.
dc.description.degreelevel	Pregrado	spa
dc.description.degreename	Profesional en estadística	spa
dc.format.mimetype	application/pdf
dc.identifier.citation	Cleves Leguízamo, D. A. (2025). Datos sintéticos: Introducción a técnicas generativas y evaluación de calidad. [Trabajo de Grado, Universidad Santo Tomás]. Repositorio Institucional
dc.identifier.instname	instname:Universidad Santo Tomás	spa
dc.identifier.reponame	reponame:Repositorio Institucional Universidad Santo Tomás	spa
dc.identifier.repourl	repourl:https://repository.usta.edu.co	spa
dc.identifier.uri	http://hdl.handle.net/11634/71463
dc.language.iso	spa
dc.publisher	Universidad Santo Tomás	spa
dc.publisher.branch	CRAI-USTA Bogotá
dc.publisher.faculty	Facultad de estadística	spa
dc.publisher.program	Rregrado estadística	spa
dc.relation.references	Abowd, J., & Lane, J. (2003). Synthetic Data and Confidentiality Protection. https://www2.census.gov/ces/tp/tp-2003-10.pdf
dc.relation.references	APX. (2025). Propensity Score Evaluation. Evaluating Synthetic Data Quality. https://apxml.com/courses/evaluating-synthetic-data-quality/chapter-2-advanced-statistical- fidelity/propensity-score-evaluation
dc.relation.references	Colombi, K. (2025). How to Generate Synthetic Data: A Comprehensive Guide. Tonic.ai. https://www.tonic.ai/guides/how-to-generate-synthetic-data-a-comprehensive-guide
dc.relation.references	DataCebo. (2025). sdmetrics. https://docs.sdv.dev/sdmetrics
dc.relation.references	Desfontaines, D. (2023). The Fundamental Trilemma of Synthetic Data Generation. https://www.tmlt.io/resources/fundamental-trilemma-synthetic-data-generation
dc.relation.references	El Namaki, M. S. S. (2025). AI Data Syndrome: Could Synthetic Data Lead to Hallucinative AI Analytics Outcomes?. https://ijeber.com/uploads2025/ijeber_05_325.pdf
dc.relation.references	Farhadyar, K., Bonofiglio, F., Hackenberg, M., Behrens, M., Zöller, D., & Binder, H. (2024). Combining Propensity Score Methods with Variational Autoencoders for Generating Synthetic Data in Presence of Latent Sub-Groups. BMC Medical Research Methodology, 24, Article 198. https://doi.org/10.1186/s12874-024-02327-x
dc.relation.references	Genest, C., & Favre, A. (1986). Everything You Always Wanted to Know About Copula Modeling but Were Afraid to Ask. https://www.uni-muenster.de/Physik.TP/ lemm/seminarSS08/JHE-2007.pdf
dc.relation.references	Hamming, R. (1986). You and Your Research. https://docs.sdv.dev/sdmetrics
dc.relation.references	Jordan, J., Szpruch, L., Houssiau, F., Bottarelli, M., Cherubin, G., Maple, C., Cohen, S. N., & Weller, A. (2024). Synthetic data: What, Why and How?. https://www.tmlt.io/resources/fundamental-trilemma-synthetic-data-generation
dc.relation.references	Kooistra, E. (2024). How do I Know that the Synthetic Data is of the Right Quality for My Use Case? https://bluegen.ai/how-do-i-know-that-the-synthetic-data-is-of-the-right-quality-for- my-use-case/
dc.relation.references	Locowic, L., Monteverdi, A., & Mendoza, E. (2024). Synthetic Data Generation From Real Data Sources Using Monte Carlo Tree Search and Large Language Models. https://www.authorea.com/users/832345/articles/1225643
dc.relation.references	Mohapatra, S., Zong, J., Kerschbaum, F., & He, X. (2022). Differentially Private Data Generation with Missing Mata. Proceedings of the VLDB Endowment. https://www.vldb.org/pvldb/vol17/p2022-mohapatra.pdf
dc.relation.references	Office of the Privacy Commissioner of Canada. (2022). When What is Old is New Again – The Reality of Synthetic Data. Privacy Tech-Know Blog. https://www.priv.gc.ca/en/blog/20221012
dc.relation.references	Pathare, A., Mangrulkar, R., Suvarna, K., Parekh, A., Thakur, G., & Gawade, A. (2023). Comparison of Tabular Synthetic Data Generation Techniques Using Propensity and Cluster Log Metric. International Journal of Intelligent Systems and Applications in Engineering, 11(3), 100177. https://www.sciencedirect.com/science/article/pii/S2667096823000241
dc.relation.references	Ravn, L. (2024). The Overlooked Politics of Synthetic Data Performance Metrics. https://www2.census.gov/ces/tp/tp-2003-10.pdf
dc.relation.references	Restrepo Lopera, J. P. (2023). Nonparametric Generation of Synthetic Data Using Copulas. https://repository.eafit.edu.co/entities/publication/e87d24db-ff18-44e6-b8b3-8eb43b1870fa
dc.relation.references	Rubin, D. B. (1993). Discussion: Statistical Disclosure Limitation. Journal of Official Statistics. https://www.scb.se/contentassets/ca21efb41fee47d293bbee5bf7be7fb3/discussion-statistical-disclosure-limitation2.pdf
dc.relation.references	Shannon, C. E. (1948). A Mathematical Theory of Information. Bell System Technical Journal, 27(3), 379–423. https://people.math.harvard.edu/ ctm/home/text/others/shannon/entropy/entropy.pdf
dc.relation.references	Snoke, J., Raab, G. M., Nowok, B., Dibben, C., & Slavkovic, A. (2018). General and specific utility measures for synthetic data. Journal of the Royal Statistical Society: Series A (Statistics in Society), 181(3), 663–688. https://academic.oup.com/jrsssa/article/181/3/663/7072005
dc.relation.references	UNESCO. (2025). Ethics of Artificial Intelligence. https://www.unesco.org/en/artificial-intelligence/recommendation-ethics
dc.rights	Attribution-NonCommercial-NoDerivs 2.5 Colombia	en
dc.rights.accessrights	info:eu-repo/semantics/openAccess
dc.rights.coar	http://purl.org/coar/access_right/c_abf2
dc.rights.local	Abierto (Texto Completo)	spa
dc.rights.local	Abierto (Texto Completo)	spa
dc.rights.uri	http://creativecommons.org/licenses/by-nc-nd/2.5/co/
dc.subject.keyword	Synthesis
dc.subject.keyword	Similarity
dc.subject.keyword	Propensity
dc.subject.keyword	Dependence
dc.subject.keyword	Copula
dc.subject.lemb	Estadísticas
dc.subject.lemb	Modelos estadísticos
dc.subject.lemb	Análisis comparativo
dc.subject.proposal	Sintetización
dc.subject.proposal	Semejanza
dc.subject.proposal	Propensión
dc.subject.proposal	Dependencia
dc.subject.proposal	Cópula
dc.title	Datos sintéticos: Introducción a técnicas generativas y evaluación de calidad
dc.type	bachelor thesis
dc.type.coar	http://purl.org/coar/resource_type/c_7a1f
dc.type.coarversion	http://purl.org/coar/version/c_ab4af688f83e57aa
dc.type.drive	info:eu-repo/semantics/bachelorThesis
dc.type.version	info:eu-repo/semantics/acceptedVersion