Lectures about secure multi-party computation (MPC)

Lecture about cryptographic technology

Thijs Veugen is speaking during the Florence Nightingale Colloquium on March 12 from 16.00 – 17.00h in Leiden. Thijs: “Within the lecture I‘ll introduce the concept of secure multi-party computation, and how it enables ‘securely analyzing shared sensitive data without sharing it’. After explaining the technology and a couple of applications that we’ve worked on, I’ll zoom into the added value for data science. Finally, I’ll sketch the maturity of this innovative cryptographic technology, and how it can be used in practice.”

Lecture about fraud detection

Thijs Veugen and Thomas Attema are speakers during the SCRIPTS Workshop on February 13 in Singapore. SCRIPTS means ‘Strategic Centre for Research on Privacy-Preserving Technologies and Systems’. The mission of SCRIPTS is for the benefits of the economy and its people. How to make the best use of collected data by having the ability to preserve the privacy and confidentiality of data, during data mining, analysis and sharing.

Alex Sangers is speaking about the same subject on February 11 at Financial Cryptography 2019 in St. Kitts.

The subject of the lecture is: ‘Secure multi-party PageRank algorithm for collaborative fraud detection’. Collaboration between financial institutions helps to improve detection of fraud. However, exchange of relevant data between these institutions is often not possible due to privacy constraints and data confidentiality.

An important example of relevant data for fraud detection is given by a transaction graph, where the nodes represent bank accounts and the links consist of the transactions between these accounts. Securely computing the PageRank values of all nodes can be used to improve fraud detection.

The execution time of our MPC implementation scales linearly with the number of nodes, and the method is highly parallelizable. Secure multi-party PageRank is feasible in a realistic setting with millions of nodes per party by extrapolating the results from our experiments.

Lecture about privacy-preserving systems

The subject of the lecture is: ‘A new approach to privacy-preserving clinical decision support systems’.

Clinical decision support systems (CDSS) are a category of health information technologies that can assist clinicians to choose optimal treatments. These support systems are based on clinical trials and expert knowledge; however, the amount of data available to these systems is limited. For this reason, CDSSs could be significantly improved by using the knowledge obtained by treating patients. This knowledge is mainly contained in patient records, whose usage is restricted due to privacy and confidentiality constraints.

We use MPC to build a secure CDSS, without burdening clinicians with the computational and communication costs. Our system is able to compute the effectiveness of 100 treatments for a specific patient in less than 24 minutes, querying a database containing 20,000 patient records.