Differential Privacy (DP) has become one of the most used approaches to protect individual data. However, its implementation can vary significantly depending on the context we are using it. In this study, we aim to compare two such implementations of DP: Google's Differential Privacy, an open-source library used for structured data analytics, and Differentially Private Offsite Prompt Tuning (DP-OPT), a tool used for the adaptation of machine learning models.

The main research question of this study is the following: "How do DP-OPT and Google's Differential Privacy Library compare when accounting for different factors in different contexts?" We aim to conduct this research by doing a literature-based study where no empirical experiments will be performed. This is because of the underlying complexity of both tools, especially DP-OPT, and the time constraints posed in this project.

The main results of this study show that Google DP is a tool that benefits from its interpretability and speed when processing data analysis, while DP-OPT shows a higher accuracy when training large language models (LLMs). Meaning that there is no single mechanism for each problem, but instead it depends on factors like the underlying task complexity, current available resources, and the final goal of the task.

By comparing these tools side-by-side, this research aims to provide more insights into how each tool behaves and performs under different contexts and tasks. We aim to guide developers and researchers to make better and more informed decisions when choosing a tool for their desired task.

Secure Multi-Party Computation (SMPC) is a widely-used cryptographic tool for privacy-preserving data analysis. The progress in the field of quantum computing has led to the development of Quantum SMPC (QSMPC), which promises informationtheoretic security based on physics laws. There is a significant number of proposed protocols. However, as the number of theoretical protocols grows, their practical viability remains unclear. Our paper presents a systematic literature review of 37 recent QSMPC protocols to assess the state of the field, focusing on protocol functionality, quantum resource requirements, privacy-ensuring techniques, and feasibility on current technologies. Our analysis reveals a plethora of innovative theoretically-private protocols, constrained by significant practical issues. We found that protocols tend to rely on a core set of quantum resources, including entangled states (Bell, GHZ), Quantum Key Distribution, and decoy particles to ensure security. We also observe a frequent reliance on a semi-honest third party. Most importantly, we identify a noticeable gap in feasibility. 19 out of 37 of the papers surveyed provide no clear discussion on the practicality of the proposed protocols. The papers that do discuss this topic describe small-scale simulations, which may require algorithmic compromises and are tested in ideal conditions only. Our findings highlight a gap between theory and practice, suggesting that real-world application of the protocols is not yet possible, given the current state of quantum hardware.

Secure multi-party computation (SMPC) is a cryptographic technique that enables multiple parties to work together on data without sharing their private information with each other. This paper investigates how two open-source frameworks, SecretFlow and FATE, implement SMPC and other privacy-preserving techniques and evaluates their usability and scalability in applied settings. The goal is to better understand the trade-offs these frameworks offer for researchers and developers seeking to build privacy-preserving machine learning pipelines. Using a small-scale experimental setup, both frameworks were deployed in controlled environments and evaluated based on setup complexity, documentation quality, modularity, architecture, and the secure computation methods used. The results show that SecretFlow offers greater SMPC flexibility, modularity, and developer usability, making it well-suited for research and rapid prototyping. FATE, while more complex to integrate, provides comprehensive workflow coordination and is better suited for production environments. Experiments were conducted on a single physical node with a small dataset, enabling reproducibility but limiting generalizability to large-scale, real-world deployments. This work provides practical insights into the usability and architectural design of the frameworks, along with a protocol-level comparison to guide developers in selecting tools for privacy-sensitive machine learning.

As financial institutions adopt more sophisticated Anti-Money Laundering (AML) techniques, such as the deployment of Graph Neural Networks (GNNs) to detect patterns, laundering behavior is likely to evolve. In this paper, we present a novel perturbation framework that models laundering as an evasion-based, restricted black-box process. Our tool systematically alters labeled laundering subgraphs through a set of parameterized graph actions (intermediary injection, merging, and splitting) designed to simulate realistic laundering adaptations. We apply our framework to one of the AMLWorld synthetic transaction datasets to generate multiple perturbed versions defined by a set of parameterized preset configuration files. We then evaluate the impact of these perturbations on two MEGA-GNN variants of the current state-of-the-art in temporal multigraph-compatible GNN architectures. Our results show that realistic structural perturbations can impact performance and serve as a valuable tool to evaluate model adaptability and robustness. Our work aims to contribute to a deeper understanding of the evolutionary dynamics between AML systems and laundering behavior.

Homomorphic Encryption (HE) enables computation directly on encrypted data, while offering strong cryptographic and privacy guarantees for data-driven sectors like healthcare, finance, and cloud computing. However, practical adoption of HE is severely limited by its computational overhead and specialized expertise requirements. This thesis investigates the financial feasibility of HE by analyzing its Total Cost of Ownership (TCO), an overlooked but crucial factor for researchers, companies, and cloud service customers deciding whether to adopt HE. Building on performance benchmarks from Guillot et al. and Lo et al., we evaluate the runtime, resource requirements, and personnel costs of Fully Homomorphic Encryption (FHE) across different workloads, and Amazon Web Service (AWS) cloud configurations. Our results show that encryption and expert labor dominate the cost, with encryption accounting for over 80% of the runtime and personnel expertise exceeding 98% of the total annual cost ($125,558.066). These findings highlight that, despite the security benefits of HE, its financial overhead restricts its adoption to organizations with substantial cryptographic expertise and budgets. This thesis equips researchers and companies with data-driven insight into HE's true economic impact and underlines the need for further optimization and automation to make HE accessible and cost-effective for broader, real-world use.

Financial institutions have a large responsibility when it comes to detecting and preventing financial crime. However, dedicated tools to aid in financial crime detection have more demand than supply. The combination of regulatory restrictions with regards to sharing client information between financial institutions and a lack of dedicated tools for financial crime detection results in a flawed system that allows criminals to evade detection and easily continue their activities by moving between institutions. This paper answers the question: How can privacy-preserving data sharing methods enable collaborative detection of malicious clients among financial institutions? Multi-Party Private Set Intersection (MPSI) allows multiple parties to intersect their respective datasets, without revealing any data to the other parties that are not in the intersection. A special case of MPSI is Threshold Multi-Party Private Set Intersection (T-MPSI), where given a threshold T, an item is only included if T or more parties hold that item. This paper implements a new version, Flagged Threshold Private Set Intersection (FT-MPSI), that adds a label to each item, where the label indicates if the client has been flagged as malicious - accused or convicted of financial crime. To be included in the intersection, the item must now also be identified by at least one party as malicious. The final result of the intersection is revealed to the computing party and can be shared with the parties holding the original items while no other information is leaked. The runtime performance of the FT-MPSI protocol is compared to that of the T-MPSI protocol. FT-MPSI is slower by a constant factor of approximately 2, compared to T-MPSI, it scales linearly to the number of parties and size of the sets of the input. FT-MPSI is a practical solution for financial institutions to use in financial crime detection.

There is an increasing need for financial institutions to be able to detect illicit activities such as money laundering. While these institutions currently rely on graph-based analytics or machine learning algorithms for such detection, inter-bank collaboration is hindered by privacy concerns and regulations. In this paper, we introduce a new protocol for computing simple fundamental graph features (specifically fan-in and fan-out degrees) directly on encrypted transaction data using the advantages of additive homomorphic encryption schemes, especially the Paillier cryptosystem. Our algorithm allows a semi-trusted third party to perform computations without accessing plaintext data, enabling privacy-preserving collaboration between banks. Through the paper, we detail the protocol design, analyze its complexity, security and correctness, and demonstrate how it reduces the gap between utility and privacy. While the protocol currently supports only basic graph metrics and assumes a common normalized currency, it offers a scalable and practical foundation for future privacy-preserving financial crime analytics.

Financial crime represents a growing issue which contemporary society is facing, especially in the form of money laundering, which aims to conceal the origin of illicit funds through a network of intermediate transactions. State of the art solutions for detection of money laundering in graph representations of financial data include supervised techniques such as Graph Neural Networks. Although provably efficient, their main limitation stands in the fact that they require a dataset of correctly and completely labeled transactions, which is often unfeasible to obtain. This work aims to explore money flow statistics as an unsupervised approach to money laundering detection, through computing statistics of accounts based on the amount of money received and sent in a certain time frame or network flow analysis using maximum flow algorithms. Therefore, this paper aims to answer two questions, namely What are the existing solutions using money flow statistics? and How would these money flow statistics methods perform on a realistic dataset of transactions?. The analysis benchmarks the identified algorithms on a realistic dataset of financial transactions in order to observe their limitations and suggest further research into how these limitations can be overcome in order to make money flow statistics methods a feasible solution to money laundering detection.

To compute something securely is to do so in a way that does not reveal (some of) the inputs, intermediate values, or outputs, to certain predetermined parties. For example, a hospital might outsource the computation of patients’ medical analyses to the cloud without the cloud provider being able to extract sensitive medical information. Researchers have proposed cryptographic protocols capable of performing secure computation for any function imaginable. However, there are still technical obstacles that hinder practical deployments of secure computation protocols. In this thesis, we identify four such impracticalities and propose techniques to address them.

A crucial building block that forms the basis of all protocols in this thesis is homomorphic encryption. Like ‘regular’ encryption it protects the encrypted values from being seen. However, it also allows one to perform computations on these encrypted values, without decrypting them. We distinguish between partially-homomorphic encryption schemes, which allow for some specific computations to be performed, and fully-homomorphic encryption schemes, which can perform any computation imaginable....

In a world of increasing threats from monopolies and oligarchies, people are increasingly looking for ways to protect their privacy. Isolating oneself from the world may be tempting, but there is a collective benefit to the processing of sensitive data. For example, hospitals use patient data to improve treatments, energy companies use power consumption data to predict grid usage, and governments can address inequality only if they measure it.

Privacy-enhancing technologies promise to close the apparent gap between privacy and utility. They provide a cryptographic solution by which statistics can be calculated without exposing individual inputs. In the world envisioned here, people gain the benefits of sharing data without exposing themselves to potential abuse.

Though solutions to societal problems are rarely if ever purely technical, this dissertation is concerned only with the technical. Specifically, with privacy-preserving summation, a protocol allowing users to learn the sum of their inputs without anyone learning the individual value of anyone else. While it may sound restrictive to focus only on summation, this is sufficient to achieve complex operations including principal component analysis, singular-value decomposition, and decision tree classifications.

In this dissertation, I provide novel methods for enforcing input and output validity in privacy-preserving summation, describe how running multiple summations in parallel leads to reconstruction attacks, and propose and evaluate countermeasures based on distributed short-cycle removal.

Validation of inputs and outputs is enforced through extensions, which can be added to any privacy-preserving summation scheme without sacrificing confidentiality. The first extension ensures that the individual pieces of data being summed over each fall within a specified numeric range. The second extension allows others to ensure that the sum published by the aggregator actually corresponds to the inputs.

Reconstruction attacks are an inherent risk when multiple summations run in sequence, regardless of the implementation of the summation protocol. When users obtain the sum A + B, one cannot learn either A or B due to the summation's privacy-preserving guarantees. However, if users subsequently also learn A + B + C, then anyone can infer C from the difference of the sums.

Understanding how and when reconstruction is possible is not trivial, especially as the numbers of variables and equations grows large. In this dissertation, I show that representing summations as a graph reveals that reconstruction coincides with the graph's cycles. In other words, removing cycles prevents reconstruction attacks. Therefore, I propose a decentralised protocol for removing short cycles. Finally, I evaluate the impact that restricting valid summation has on distributed averaging, and find that though the effect is largely negative, this can mostly be ameliorated through a subsequent greedy repair algorithm.

A substantial number of passengers in Europe rely on trains for transportation, facilitated by a network of high-speed international trains. However, the coordination of train schedules across multiple networks often poses challenges due to incompatible timings. The scheduling of multiple train networks shares similarities with multi-processor task scheduling and airline scheduling but is distinguished by its cooperative nature rather than a competitive one. Cooperative scheduling necessitates the sharing of private information. This information, 'demand', is commercial sensitive information, since it can reveal demographic information like incomes and tax returns. Privacy-preserving protocols can enable the computation of statistics without revealing this demands (in railway systems) to unauthorized parties. Despite the critical role of privacy in multi-party scheduling, research in this domain remains limited due to domain specific constraints. A model supporting such privacy considerations could significantly help Europe achieve its carbon-neutral goals while improving cross-border services. In this research, we propose a system designed to facilitate joint service scheduling, ensuring confidentiality, integrity, and authenticity. We use partial and fully homomorphic encryption techniques that mimic the outcomes achievable with a trusted third party. We conduct a comparative analysis of online and offline approaches, emphasizing how they achieve confidentiality, collusion-resistance, traceability and non-repudiation. Theoretical and experimental evaluations demonstrate the feasibility of the system for real-world applications by creating schedules for upto four parties. Our solution for scheduling seven slots takes approximately three hours, which is a feasible duration to solve a problem of this size.

Within the field of \emph{cyber threat intelligence} (CTI), healthcare institutions are one of the most targeted organizations by cybercriminals. To mitigate future attacks on their digital infrastructures, healthcare institutions can collaborate and exchange security logs. These logs include data such as IP addresses, malware hashes, and other indicators of compromise. By identifying shared elements across different datasets, threats that are harmful to a greater number of organizations---and thus, pose a more significant risk---can be highlighted to detect common attack patterns. These attack patterns could then provide insight into understanding how cyber criminals operate on a larger scale. However, disclosing locally collected CTI could compromise a hospital's security posture and reputation since it reveals vulnerabilities or attack techniques used by hackers. Furthermore, this cyber threat data is often sensitive. That is why the threat data needs to be shared in a privacy-preserving manner. \emph{Multiparty private set intersection} (MPSI) is a solution that allows parties to find the intersection of all their sets without learning anything of the other inputs. Although, in many cases, the condition that an element be present in all sets is too restrictive. A potential threat is still worth investigating even if it only appears in a portion of sets. Therefore, we focus on \emph{threshold multiparty private set intersection} (T-MPSI). However, not all T-MPSI schemes perform equally depending on the context. Our goal is to determine what makes a T-MPSI protocol effective for application in medical CTI sharing. To do so, we analyze four state-of-the-art T-MPSI protocols in terms of security, theoretical communication and computational complexities, and practical runtime performance.

Financial crime has seen a surge in complexity over the past decades as a result of digitisation. This required better tools for detecting financial crime, creating a cat-and-mouse game between law enforcement and criminals. Anti-money laundering (AML) is the collective term describing these tools, as well as financial regulations designed to combat money laundering. The detection of fraud is done through data analysis. Typically, the financial data is a large set of transactions between accounts. Transactions have a sender, receiver, and an amount that is transferred from sender to receiver. This can be represented by a directed graph where nodes represent accounts and edges transactions between accounts. Methods for finding fraudulent transactions in transactional graphs are widely studied in both literature and industry. This usually assumes a centralised entity with full access to the data. At an international scale, the centralised approach is not suitable. Besides its massive scale posing a challenge, it is hard to imagine a global agreement on what capabilities such an entity should have. At the same time, organisations sharing financial data for analysis raises privacy concerns. We limit our scope to transactional cycles. These typically occur when criminals send money through a series of international bank accounts back to the original account. This lowers the chance of the money being traced back to its illicit source. We propose a privacy-preserving decentralised protocol which takes as input a parameter $\ell$ from any node and outputs all cycles of at most length $\ell$ containing the node. The protocol relies on the decisional Diffie-Hellman assumption, and we show the protocol is secure against a polynomially bounded adversary within our security model. The communication complexity of an instance is $O(n^{\ell})$, multiple instances can be ran in parallel. We implement the protocol and measure its performance on scale free graphs in terms of communication and computational costs. Lastly, we discuss our results and suggestions for future work.

Ensuring the privacy of medical data in a meaningful manner is a complex task. This domain presents a plethora of unique challenges: high stakes, vast differences between possible use cases, long-established methods that limit the number of feasible solutions, and more. Consequently, an effective approach to ensuring the privacy of medical data must be easy to adopt, offer robust privacy guarantees, and minimize the reduction in data utility.

The unique nature of medical data presents distinct challenges and also opportunities. We consider various types of correlations that significantly impact privacy guarantees. However, these correlations can also be used to train a model for removing anomalies and subsequently enhancing the utility of synthetic medical data.

This thesis proposes a framework compatible with state-of-the-art approaches for differentially private dataset release based on the usage of Generative Adversarial Networks (GANs). Our framework uses a part of the privacy budget to train an unsupervised learning model to detect and remove anomalies. We evaluate the performance of the framework using a variety of machine-learning models and metrics. The final results show an improvement of up 13% compared to approaches not using our framework, under the same privacy budget.

Digital modelling is becoming more prevalent in many applications. The underlying 3D mesh objects are therefore getting increasingly valuable, such that methods for ownership protection are required. Watermarking is a solution to this problem, yet the research combining watermarking and 3D mesh objects is limited. This work aims to improve an existing watermarking method that embeds watermarking bits into a 3D model by selecting features, or the vertices with a larger mean curvature. We propose an improvement where a centre of mass calculation is changed to a centre of volume, aiming to achieve better robustness against mesh simplification attacks on 3D mesh objects by use of this volume centre instead of a mass centre. The proposed method of applying the watermark reaches an accuracy score that is 15.78 percentage points higher on average than the original method, when attacked by mesh simplification. Therefore, this work shows a more robust approach to the method build upon.

In the digital era, XML data is fundamental for various applications, requiring robust methods to ensure data integrity and security. Traditional digital watermarking techniques face challenges due to XML's hierarchical structure. Zero-watermarking, which derives a watermark from inherent data properties without altering the data, offers a promising alternative. This paper enhances zero-watermarking robustness for XML by integrating usability queries to filter functional dependencies (FDs) based on their relevance. This approach maintains watermark integrity even when some data elements are modified, provided the data's usability is preserved. The proposed method increases resilience against targeted attacks altering semantic information while preserving usability. Key contributions include incorporating usability queries into the FD-based zero-watermarking process, generating cover ranges for usability queries, and empirically testing the method's robustness.

AI and machine learning have been topics of big interest in the last couple of years, with plenty of applications in many domains. To train these models into useful and desirable tools, a large amount of data is necessary. This data is expensive to collect, becoming one of the most valuable commodities of this century. As the value of data increases, protecting this intellectual property becomes more and more relevant. Watermarking is a technique widely used for data protection in media, but the non-media counterpart has not been researched as thoroughly. In this paper, an adaptation of a common watermarking technique, DWT watermarking, is applied on two datasets used for machine learning. This technique is invisible and robust in signal watermarking, but its performance on a numerical dataset has not been previously researched. A previously devised algorithm was used, but it was adjusted to better fit dataset watermarking. To assess the quality of the watermark, the marked data has been subjected to create, remove, update and zero-out attacks. On top of this, multiple machine-learning models have been trained on the marked data. Initial results show that the proposed technique performs well in terms of invisibility, obtaining similar or better accuracies than models trained on the original data, but it is quite sensitive to attacks. Even small modifications, less than 1\% of the data, can break the signature.