An Approach to Trade-off Privacy and Classification Accuracy in Machine Learning Processes

Aug 4, 2026

Upload by Ophelia

Machine learning over large, distributed data archives can expose sensitive information: even when data are protected, high-capacity models may still enable identification of individuals. This discussion paper introduces a decision-support framework for data anonymization that uses relaxed functional dependencies (rfds) and data correlations to discover anonymization strategies. The framework supports simultaneous exploitation of multiple correlations, selection of strategies, and clear evaluation of trade-offs between privacy and data utility. Experiments on real datasets show promising utility while meeting privacy requirements.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages11

Reading Time28 min

AccessFree to download

Introduction

Privacy challenges in Big Data analytics

Abstracted framework and decision support

Correlation-based anonymization using relaxed functional dependencies (rfds)

Trade-off analysis between privacy and data utility

Experimental evaluation

Results on real-life datasets

An Approach to Trade-off Privacy and Classification Accuracy in Machine Learning Processes Citation for published version (APA): Caruccio, L. , Desiato, D. , Polese, G. , Tortora, G. , & Zannone, N. (2023) . An Approach to Trade-off Privacy and Classification Accuracy in Machine Learning Processes. In D. Calvanese, C. Diamantini, G. Faggioli, N. Ferro, S. Marchesin, G. Silvello, & L. Tanca (Eds.), SEBD 2023 : 31st Symposium of Advanced Database Systems: Proceedings of the 31st Symposium of Advanced Database Systems : Galzingano Terme, Italy, July 2nd to 5th, 2023 (pp. 420-429) . (CEUR Workshop Proceedings; Vol. 3478) . [CEUR-WS.org](CEUR-WS.org). [https://ceur-ws.org/Vol-](https://ceur-ws.org/Vol-)[ ](https://ceur-ws.org/Vol-)[3478/paper72.pdf](3478/paper72.pdf) Document license: CC BY Document status and date: Published: 01/01/2023 Document Version: Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers) Please check the document version of this publication: • A submitted manuscript is the version of the article upon submission and before peer-review. There can be important differences between the submitted version and the official published version of record. People interested in the research are advised to contact the author for the final version of the publication, or visit the DOI to the publisher's website. • The final author version and the galley proof are versions of the publication after peer review. • The final published version features the final layout of the paper including the volume, issue and page numbers. Link to publication General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal. If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license above, please follow below link for the End User Agreement: [www.tue.nl/taverne](www.tue.nl/taverne) Take down policy If you believe that this document breaches copyright please contact us at: [openaccess@tue.nl](openaccess@tue.nl) providing details and we will investigate your claim. Download date: 28. Apr. 2026 An Approach to Trade-off Privacy and Classification Accuracy in Machine Learning Processes Loredana Caruccio1 , Domenico Desiato2 , Giuseppe Polese1 , Genoveffa Tortora1 and Nicola Zannone3 1 Department of Computer Science, University of Salerno, via Giovanni Paolo II n. 132, 84084 Fisciano (SA), Italy 2 Department of Computer Science, University of Bari Aldo Moro, via Edoardo Orabona n.4, 70125 Bari (BA), Italy 3 Eindhoven University of Technology, Eindhoven, Netherlands Abstract Machine learning techniques applied to large and distributed data archives might result in the disclosure of sensitive information. Data often contain sensitive identifiable information, and even if these are protected, the excessive processing capabilities of current machine learning techniques might facilitate the identification of individuals. This discussion paper presents a decision-support framework for data anonymization. The latter relies on a novel approach that exploits data correlations, expressed in terms of relaxed functional dependencies (rfds), to identify data anonymization strategies for providing suitable trade-offs between privacy and data utility. It also permits to generate anonymization strategies leveraging multiple data correlations simultaneously t

FAQs

How does the framework help users compare privacy vs. data utility?

It enables understanding of the trade-offs offered by obtained anonymization strategies, supporting selection of strategies that meet both privacy and utility requirements. It also can generate strategies leveraging multiple correlations simultaneously.

Aug 4, 2026

Upload by Ophelia