DescriptionData marketplaces are online platforms that provide a way for individuals to monetize their (personal) data. In traditional data marketplaces, the data is uploaded to the marketplace platform in plain-text. Since the broker has access to all data, such marketplaces are risk to the users' privacy. We present a privacy-preserving marketplace that allows data owners to keep control over their data. We use secure multi-party computation to enable data consumers to evaluate expressive functions on a set of data. In our marketplace, the broker has neither access to the data nor the results of a function evaluation. Furthermore, our design ensures data-origin authenticity and enables data owners to define data-usage policies that are enforced by the computation nodes. In addition to the architecture of this private marketplace, we discuss a reference implementation. We also provide an evaluation of our approach, demonstrating its practicability.
|Period||9 Dec 2022|
|Event title||2022 International Conference on Emerging Networking Experiments and Technologies: CoNEXT 2022|
|Degree of Recognition||International|
Research output: Chapter in Book/Report/Conference proceeding › Conference paper › peer-review
Project: Research project