Master theses

Introduction

The growing concern for data privacy and ownership has led to an increased interest in decentralized data storage solutions. Data spaces offer a way for individuals to retain control over their personal information by storing it in personal data pods or vaults. This approach aligns with emerging privacy regulations and enhances data interoperability across services. This thesis proposes to refactor the open-source Elevate sport app to leverage data spaces, allowing users to store their fitness data securely in their own pods. The Elevate app, designed to track and analyze training progress for various sports, currently relies on a centralized data model. Transitioning to a decentralized model aims to improve data privacy, interoperability, and app performance.

Methodology

The first phase, data transformation, will focus on converting existing data formats, such as .fit files, into semantically annotated data that can be stored in decentralized data pods using RML. This will involve mapping the existing data into RDF-based Linked Data formats to ensure compatibility with the Solid specification. By adopting standardized semantic annotations, the data can be easily queried and shared across different services while maintaining privacy and interoperability. The second phase, refactoring and integration, will involve redesigning the data access layer of the Elevate app to interact directly with user-managed data pods. The app will need to be modified to use the Solid protocol to access the users data. An example of an existing application using this technologie is the Solid Watchparty. In the final phase, performance optimization with aggregators, the focus will be on improving data retrieval efficiency and reducing loading times. Aggregators will be integrated to manage decentralized data queries by maintaining up-to-date, materialized views. This approach will minimize the time required to access and process data by allowing the Elevate app to fetch pre-computed results directly from aggregators. Throughout these phases, the student will gain hands-on experience in front-end development, web technologies, data spaces, and semantic web technologies. By the end of the project, the refactored Elevate app is expected to demonstrate the practical benefits of data spaces and aggregators, providing a scalable and privacy-focused solution for sport-tracking applications.

The objective of this thesis is to refactor the Elevate sport app to a decentralized data model using data spaces and aggregators, aiming to improve both data privacy and app performance. By storing user data in personal pods according to the Solid specification, users can retain control over their own information, while the app can handle data retrieval more efficiently.

Additionally, this project seeks to significantly reduce the app's loading times by using aggregators that maintain up-to-date and semantically enriched views of decentralized data. This will demonstrate that decentralized storage and data processing are not only feasible but also beneficial for sport-tracking applications.

Throughout this project, the student will develop essential skills in front-end development, semantic web technologies, and data privacy, with a strong focus on innovation and research into decentralized systems.

The main objectives of this thesis are:

1. Refactor Elevate to Support Data Spaces: Adapt the app to store user data in personal pods using the Solid specification, ensuring secure and private data management.
2. Semantic Data Transformation: Convert existing .fit files and other data formats into semantically annotated data that can be stored and queried in data pods.
3. Performance Optimization with Aggregators: Explore the use of aggregators to reduce loading times and manage decentralized data queries efficiently.
4. Skill Development: Enable the student to gain expertise in front-end development, web technologies, data spaces, and semantic web technologies.