arXiv subjects

Elias Grünewald

Publications and source records attributed to Elias Grünewald.

17 recordsLinked to original sources

arXiv2026

An Event-Driven Cloud-Native Wearable Analytics Framework for Real-Time Clinical Workloads

Continuous physiological monitoring using consumer-grade wearables offers a transformative opportunity for clinical care and research, yet integration remains hindered by device heterogeneity, proprietary data formats, and strict regulatory requirements. We present an event-driven, cloud-native system designed to ingest, normalize, and analyze high-frequency vital signs from wearables at scale and without vendor lock-in. The system design proposes a multi-layered microservice architecture using cluster orchestration. Data acquisition is handled via a cross-platform mobile application that leverages native health frameworks, ensuring compatibility across fragmented device ecosystems. To address interoperability, we implement an event-driven transformation pipeline using stream processing engines and specialized services to map raw measurements to the FHIR standard for medical interoperability. Our novel dependency-aware FHIR minimization scheme reduces storage overhead while maintaining lossless resource reconstruction. Furthermore, the platform integrates a modular data analytics and machine learning layer based on a medallion lakehouse architecture, supporting the full machine learning lifecycle from real-time stream processing to model serving. Performance evaluation demonstrates that the ingestion pipeline sustains 50 full ingestion requests per second with median response times under 8 ms, satisfying the low-latency requirements for real-time patient monitoring. Our open-source implementation adheres to regulatory compliance standards through role-based access control and secure service-to-service communication, providing a robust foundation for deploying wearable-based monitoring in institutional healthcare settings for clinical decision support and research workloads.

cs.DC↗

arXiv2026

Codebase-Memory: Tree-Sitter-Based Knowledge Graphs for LLM Code Exploration via MCP

Large Language Model (LLM) coding agents typically explore codebases through repeated file-reading and grep-searching, consuming thousands of tokens per query without structural understanding. We present Codebase-Memory, an open-source system that constructs a persistent, Tree-Sitter-based knowledge graph via the Model Context Protocol (MCP), parsing 66 languages through a multi-phase pipeline with parallel worker pools, call-graph traversal, impact analysis, and community discovery. Evaluated across 31 real-world repositories, Codebase-Memory achieves 83% answer quality versus 92% for a file-exploration agent, at ten times fewer tokens and 2.1 times fewer tool calls. For graph-native queries such as hub detection and caller ranking, it matches or exceeds the explorer on 19 of 31 languages.

cs.SE↗

arXiv2025

Prink: $k_s$-Anonymization for Streaming Data in Apache Flink

In this paper, we present Prink, a novel and practically applicable concept and fully implemented prototype for ks-anonymizing data streams in real-world application architectures. Building upon the pre-existing, yet rudimentary CASTLE scheme, Prink for the first time introduces semantics-aware ks-anonymization of non-numerical (such as categorical or hierarchically generalizable) streaming data in a information loss-optimized manner. In addition, it provides native integration into Apache Flink, one of the prevailing frameworks for enterprise-grade stream data processing in numerous application domains. Our contributions excel the previously established state of the art for the privacy guarantee-providing anonymization of streaming data in that they 1) allow to include non-numerical data in the anonymization process, 2) provide discrete datapoints instead of aggregates, thereby facilitating flexible data use, 3) are applicable in real-world system contexts with minimal integration efforts, and 4) are experimentally proven to raise acceptable performance overheads and information loss in realistic settings. With these characteristics, Prink provides an anonymization approach which is practically feasible for a broad variety of real-world, enterprise-grade stream processing applications and environments.

cs.DC↗

arXiv2024

Extending Business Process Management for Regulatory Transparency

Ever-increasingly complex business processes are enabled by loosely coupled cloud-native systems. In such fast-paced development environments, data controllers face the challenge of capturing and updating all personal data processing activities due to considerable communication overhead between development teams and data protection staff. To date, established business process management methods generate valuable insights about systems, however, they do not account for all regulatory transparency obligations. For instance, data controllers need to record all information about data categories, legal purpose specifications, third-country transfers, etc. Therefore, we propose to bridge the gap between business processes and application systems by providing three contributions that assist in modeling, discovering, and checking personal data transparency through a process-oriented perspective. We enable transparency modeling for relevant business activities by providing a plug-in extension to BPMN featuring regulatory transparency information. Furthermore, we utilize event logs to record regulatory transparency information in realistic cloud-native systems. On this basis, we leverage process mining techniques to discover and analyze personal data flows in business processes, e.g., through transparency conformance checking. We design and implement prototypes for all contributions, emphasizing the appropriate integration and modeling effort required to create business-process-oriented transparency. Altogether, we connect current business process engineering techniques with regulatory needs as imposed by the GDPR and other legal frameworks.

cs.SE↗

arXiv2024

Hook-in Privacy Techniques for gRPC-based Microservice Communication

gRPC is at the heart of modern distributed system architectures. Based on HTTP/2 and Protocol Buffers, it provides highly performant, standardized, and polyglot communication across loosely coupled microservices and is increasingly preferred over REST- or GraphQL-based service APIs in practice. Despite its widespread adoption, gRPC lacks any advanced privacy techniques beyond transport encryption and basic token-based authentication. Such advanced techniques are, however, increasingly important for fulfilling regulatory requirements. For instance, anonymizing or otherwise minimizing (personal) data before responding to requests, or pre-processing data based on the purpose of the access may be crucial in certain usecases. In this paper, we therefore propose a novel approach for integrating such advanced privacy techniques into the gRPC framework in a practically viable way. Specifically, we present a general approach along with a working prototype that implements privacy techniques, such as data minimization and purpose limitation, in a configurable, extensible, and gRPC-native way utilizing a gRPC interceptor. We also showcase how to integrate this contribution into a realistic example of a food delivery use case. Alongside these implementations, a preliminary performance evaluation shows practical applicability with reasonable overheads. Altogether, we present a viable solution for integrating advanced privacy techniques into real-world gRPC-based microservice architectures, thereby facilitating regulatory compliance ``by design''.

cs.CR↗

arXiv2023

Towards Cross-Provider Analysis of Transparency Information for Data Protection

Transparency and accountability are indispensable principles for modern data protection, from both, legal and technical viewpoints. Regulations such as the GDPR, therefore, require specific transparency information to be provided including, e.g., purpose specifications, storage periods, or legal bases for personal data processing. However, it has repeatedly been shown that all too often, this information is practically hidden in legalese privacy policies, hindering data subjects from exercising their rights. This paper presents a novel approach to enable large-scale transparency information analysis across service providers, leveraging machine-readable formats and graph data science methods. More specifically, we propose a general approach for building a transparency analysis platform (TAP) that is used to identify data transfers empirically, provide evidence-based analyses of sharing clusters of more than 70 real-world data controllers, or even to simulate network dynamics using synthetic transparency information for large-scale data-sharing scenarios. We provide the general approach for advanced transparency information analysis, an open source architecture and implementation in the form of a queryable analysis platform, and versatile analysis examples. These contributions pave the way for more transparent data processing for data subjects, and evidence-based enforcement processes for data protection authorities. Future work can build upon our contributions to gain more insights into so-far hidden data-sharing practices.

cs.CY↗

arXiv2023

Hawk: DevOps-driven Transparency and Accountability in Cloud Native Systems

Transparency is one of the most important principles of modern privacy regulations, such as the GDPR or CCPA. To be compliant with such regulatory frameworks, data controllers must provide data subjects with precise information about the collection, processing, storage, and transfer of personal data. To do so, respective facts and details must be compiled and always kept up to date. In traditional, rather static system environments, this inventory (including details such as the purposes of processing or the storage duration for each system component) could be done manually. In current circumstances of agile, DevOps-driven, and cloud-native information systems engineering, however, such manual practices do not suit anymore, making it increasingly hard for data controllers to achieve regulatory compliance. To allow for proper collection and maintenance of always up-to-date transparency information smoothly integrating into DevOps practices, we herein propose a set of novel approaches explicitly tailored to specific phases of the DevOps lifecycle most relevant in matters of privacy-related transparency and accountability at runtime: Release, Operation, and Monitoring. For each of these phases, we examine the specific challenges arising in determining the details of personal data processing, develop a distinct approach and provide respective proof of concept implementations that can easily be applied in cloud native systems. We also demonstrate how these components can be integrated with each other to establish transparency information comprising design- and runtime-elements. Furthermore, our experimental evaluation indicates reasonable overheads. On this basis, data controllers can fulfill their regulatory transparency obligations in line with actual engineering practices.

cs.DC↗

arXiv2023

A Human-in-the-Loop Approach for Information Extraction from Privacy Policies under Data Scarcity

Machine-readable representations of privacy policies are door openers for a broad variety of novel privacy-enhancing and, in particular, transparency-enhancing technologies (TETs). In order to generate such representations, transparency information needs to be extracted from written privacy policies. However, respective manual annotation and extraction processes are laborious and require expert knowledge. Approaches for fully automated annotation, in turn, have so far not succeeded due to overly high error rates in the specific domain of privacy policies. In the end, a lack of properly annotated privacy policies and respective machine-readable representations persists and enduringly hinders the development and establishment of novel technical approaches fostering policy perception and data subject informedness. In this work, we present a prototype system for a `Human-in-the-Loop' approach to privacy policy annotation that integrates ML-generated suggestions and ultimately human annotation decisions. We propose an ML-based suggestion system specifically tailored to the constraint of data scarcity prevalent in the domain of privacy policy annotation. On this basis, we provide meaningful predictions to users thereby streamlining the annotation process. Additionally, we also evaluate our approach through a prototypical implementation to show that our ML-based extraction approach provides superior performance over other recently used extraction models for legal documents.

cs.CY↗

arXiv2023

Streamlining personal data access requests: From obstructive procedures to automated web workflows

Transparency and data portability are two core principles of modern privacy legislations such as the GDPR. From the regulatory perspective, providing individuals (data subjects) with access to their data is a main building block for implementing these. Different from other privacy principles and respective regulatory provisions, however, this right to data access has so far only seen marginal technical reflection. Processes related to performing data subject access requests (DSARs) are thus still to be executed manually, hindering the concept of data access from unfolding its full potential. To tackle this problem, we present an automated approach to the execution of DSARs, employing modern techniques of web automation. In particular, we propose a generic DSAR workflow model, a corresponding formal language for representing the particular workflows of different service providers (controllers), a publicly accessible and extendable workflow repository, and a browser-based execution engine, altogether providing ``one-click'' DSARs. To validate our approach and technical concepts, we examine, formalize and make publicly available the DSAR workflows of 15 widely used service providers and implement the execution engine in a publicly available browser extension. Altogether, we thereby pave the way for automated data subject access requests and lay the groundwork for a broad variety of subsequent technical means helping web users to better understand their privacy-related exposure to different service providers.

cs.CY↗

arXiv2023

Enabling Versatile Privacy Interfaces Using Machine-Readable Transparency Information

Transparency regarding the processing of personal data in online services is a necessary precondition for informed decisions on whether or not to share personal data. In this paper, we argue that privacy interfaces shall incorporate the context of display, personal preferences, and individual competences of data subjects following the principles of universal design and usable privacy. Doing so requires -- among others -- to consciously decouple the provision of transparency information from their ultimate presentation. To this end, we provide a general model of how transparency information can be provided from a data controller to data subjects, effectively leveraging machine-readable transparency information and facilitating versatile presentation interfaces. We contribute two actual implementations of said model: 1) a GDPR-aligned privacy dashboard and 2) a chatbot and virtual voice assistant enabled by conversational AI. We evaluate our model and implementations with a user study and find that these approaches provide effective and time-efficient transparency. Consequently, we illustrate how transparency can be enhanced using machine-readable transparency information and how data controllers can meet respective regulatory obligations.

cs.SE↗

arXiv2022

Scalable Discovery and Continuous Inventory of Personal Data at Rest in Cloud Native Systems

Cloud native systems are processing large amounts of personal data through numerous and possibly multi-paradigmatic data stores (e.g., relational and non-relational databases). From a privacy engineering perspective, a core challenge is to keep track of all exact locations, where personal data is being stored, as required by regulatory frameworks such as the European General Data Protection Regulation. In this paper, we present Teiresias, comprising i) a workflow pattern for scalable discovery of personal data at rest, and ii) a cloud native system architecture and open source prototype implementation of said workflow pattern. To this end, we enable a continuous inventory of personal data featuring transparency and accountability following DevOps/DevPrivOps practices. In particular, we scope version-controlled Infrastructure as Code definitions, cloud-based storages, and how to integrate the process into CI/CD pipelines. Thereafter, we provide iii) a comparative performance evaluation demonstrating both appropriate execution times for real-world settings, and a promising personal data detection accuracy outperforming existing proprietary tools in public clouds.

cs.CR↗

arXiv2022

Advancing Blockchain-based Federated Learning through Verifiable Off-chain Computations

Federated learning may be subject to both global aggregation attacks and distributed poisoning attacks. Blockchain technology along with incentive and penalty mechanisms have been suggested to counter these. In this paper, we explore verifiable off-chain computations using zero-knowledge proofs as an alternative to incentive and penalty mechanisms in blockchain-based federated learning. In our solution, learning nodes, in addition to their computational duties, act as off-chain provers submitting proofs to attest computational correctness of parameters that can be verified on the blockchain. We demonstrate and evaluate our solution through a health monitoring use case and proof-of-concept implementation leveraging the ZoKrates language and tools for smart contract-based on-chain model management. Our research introduces verifiability of correctness of learning processes, thus advancing blockchain-based federated learning.

cs.CR↗

arXiv2021

Datensouveränität für Verbraucher:innen: Technische Ansätze durch KI-basierte Transparenz und Auskunft im Kontext der DSGVO

A sufficient level of data sovereignty is extremely difficult for consumers in practice. The EU General Data Protection Regulation guarantees comprehensive data subject rights, which must be implemented by responsible controllers through technical and organizational measures. Traditional approaches, such as the provision of lengthy data protection declarations or the downloading of raw personal data without further assistance, do not meet the requirements of informational self-determination. The new technical approaches outlined below, in particular AI-based transparency and access modalities, demonstrate the practicability of effective and versatile mechanisms. For this purpose, the relevant transparency information is extracted in a semi-automated way, represented in a machine-readable format, and then played out via diverse channels such as virtual assistants or the enrichment of search results. --- Hinreichende Datensouveränität gestaltet sich für Verbraucher:innen in der Praxis als äußerst schwierig. Die Europäische Datenschutzgrundverordnung garantiert umfassende Betroffenenrechte, die von verantwortlichen Stellen durch technisch-organisatorische Maßnahmen umzusetzen sind. Traditionelle Vorgehensweisen wie die Bereitstellung länglicher Datenschutzerklärungen oder der ohne weitere Hilfestellungen angebotene Download von personenbezogenen Rohdaten werden dem Anspruch der informationellen Selbstbestimmung nicht gerecht. Die im Folgenden aufgezeigten neuen technischen Ansätze insbesondere KI-basierter Transparenz- und Auskunftsmodalitäten zeigen die Praktikabilität wirksamer und vielseitiger Mechanismen. Hierzu werden die relevanten Transparenzangaben teilautomatisiert extrahiert, maschinenlesbar repräsentiert und anschließend über diverse Kanäle wie virtuelle Assistenten oder die Anreicherung von Suchergebnissen ausgespielt.

cs.CY↗

arXiv2021

Cloud Native Privacy Engineering through DevPrivOps

Cloud native information systems engineering enables scalable and resilient service infrastructures for all major online offerings. These are built following agile development practices. At the same time, a growing demand for privacy-friendly services is articulated by societal norms and policy through effective legislative frameworks. In this paper, we identify the conceptual dimensions of cloud native privacy engineering and propose an integrative approach to be addressed in practice to overcome the shortcomings of existing privacy enhancing technologies. Furthermore, we propose a reference software development lifecycle called DevPrivOps to enhance established agile development methods with respect to privacy. Altogether, we show that cloud native privacy engineering advances the state of the art of privacy by design and by default using latest technologies.

cs.SE↗

arXiv2021

RedCASTLE: Practically Applicable $k_s$-Anonymity for IoT Streaming Data at the Edge in Node-RED

In this paper, we present RedCASTLE, a practically applicable solution for Edge-based $k_s$-anonymization of IoT streaming data in Node-RED. RedCASTLE builds upon a pre-existing, rudimentary implementation of the CASTLE algorithm and significantly extends it with functionalities indispensable for real-world IoT scenarios. In addition, RedCASTLE provides an abstraction layer for smoothly integrating $k_s$-anonymization into Node-RED, a visually programmable middleware for streaming dataflows widely used in Edge-based IoT scenarios. Last but not least, RedCASTLE also provides further capabilities for basic information reduction that complement $k_s$-anonymization in the privacy-friendly implementation of usecases involving IoT streaming data. A preliminary performance assessment finds that RedCASTLE comes with reasonable overheads and demonstrates its practical viability.

cs.CR↗

arXiv2021

TIRA: An OpenAPI Extension and Toolbox for GDPR Transparency in RESTful Architectures

Transparency - the provision of information about what personal data is collected for which purposes, how long it is stored, or to which parties it is transferred - is one of the core privacy principles underlying regulations such as the GDPR. Technical approaches for implementing transparency in practice are, however, only rarely considered. In this paper, we present a novel approach for doing so in current, RESTful application architectures and in line with prevailing agile and DevOps-driven practices. For this purpose, we introduce 1) a transparency-focused extension of OpenAPI specifications that allows individual service descriptions to be enriched with transparency-related annotations in a bottom-up fashion and 2) a set of higher-order tools for aggregating respective information across multiple, interdependent services and for coherently integrating our approach into automated CI/CD-pipelines. Together, these building blocks pave the way for providing transparency information that is more specific and at the same time better reflects the actual implementation givens within complex service architectures than current, overly broad privacy statements.

cs.SE↗

arXiv2020

TILT: A GDPR-Aligned Transparency Information Language and Toolkit for Practical Privacy Engineering

In this paper, we present TILT, a transparency information language and toolkit explicitly designed to represent and process transparency information in line with the requirements of the GDPR and allowing for a more automated and adaptive use of such information than established, legalese data protection policies do. We provide a detailed analysis of transparency obligations from the GDPR to identify the expressiveness required for a formal transparency language intended to meet respective legal requirements. In addition, we identify a set of further, non-functional requirements that need to be met to foster practical adoption in real-world (web) information systems engineering. On this basis, we specify our formal language and present a respective, fully implemented toolkit around it. We then evaluate the practical applicability of our language and toolkit and demonstrate the additional prospects it unlocks through two different use cases: a) the inter-organizational analysis of personal data-related practices allowing, for instance, to uncover data sharing networks based on explicitly announced transparency information and b) the presentation of formally represented transparency information to users through novel, more comprehensible, and potentially adaptive user interfaces, heightening data subjects' actual informedness about data-related practices and, thus, their sovereignty. Altogether, our transparency information language and toolkit allow - differently from previous work - to express transparency information in line with actual legal requirements and practices of modern (web) information systems engineering and thereby pave the way for a multitude of novel possibilities to heighten transparency and user sovereignty in practice.

cs.CY↗