Link Prediction using Graph Neural Networks for Master Data Management

TL;DR

This paper introduces novel methods for link prediction in master data management using graph neural networks, addressing privacy, explainability, and verification challenges in sensitive applications.

Contribution

It presents new techniques for anonymizing data, training models, and ensuring explainability and verification specifically for link prediction in master data management.

Findings

Effective anonymization methods developed

Enhanced explainability techniques proposed

Successful application demonstrated in privacy-sensitive contexts

Abstract

Learning graph representations of n-ary relational data has a number of real world applications like anti-money laundering, fraud detection, and customer due diligence. Contact tracing of COVID19 positive persons could also be posed as a Link Prediction problem. Predicting links between people using Graph Neural Networks requires careful ethical and privacy considerations than in domains where GNNs have typically been applied so far. We introduce novel methods for anonymizing data, model training, explainability and verification for Link Prediction in Master Data Management, and discuss our results.