Abstract
Sharing of telecommunication network data, for example, even at highaggregation levels, is nowadays highly restricted due to privacy legislationand regulations and other important ethical concerns. It leads to scatteringdata across institutions, regions, and states, inhibiting the usage of AImethods that could otherwise take advantage of data at scale. It creates theneed to build a platform to control such data, build models or performcalculations. In this work, we propose an approach to building the bridge amonganomaly detection, federated learning, and data streams. The overarching goalof the work is to detect anomalies in a federated environment over distributeddata streams. This work complements the state-of-the-art by adapting the datastream algorithms in a federated learning setting for anomaly detection and bydelivering a robust framework and demonstrating the practical feasibility in areal-world distributed deployment scenario.