ING as a European leading bank is continuously keeping track of its digital assets such as servers, network devices, and software programs. Asset management is a challenging task because of its complex dependencies and hierarchical nature. Despite these difficulties, many monitoring tools were successfully implemented in ING, including metric anomaly detection, resource capacity forecasting, IT structure optimization, and incident management with natural language processing of logs and events.

In this talk, we will discuss incident management with a Hawkes process. This model was previously successfully applied for earthquake predictions based on aftershocks and capturing the dynamics of order books in finance. The Hawkes process model is well-defined mathematically and can process a large volume of data to uncover Granger causal structures in data if implemented appropriately. We show how Hawkes processes help answer the question ‘what caused what’ within the IT infrastructure of a major European bank.

NOTE: This talk will focus mostly on explaining the mathematics behind Hawkes processes. Although we assume no prior knowledge, it might be nice to refresh on Poisson processes, which will be our starting point (although we recap them in our presentation as well).

See for instance: https://builtin.com/data-science/poisson-process