We look at the processing of stateful streaming data, state interfaces, and the implementation of state management and fault tolerance in Apache Flink.
There are multiple approaches to an unsupervised anomaly detection problem that try to exploit the differences between the properties of common and unique observations.
CORE
