State-Based Programming in PQL

Abstract

Complex Event Processing (CEP) has become increasingly popular within organizations. The financial industry used to be the sole beneficiary of the CEP capabilities, however nowadays we see it increasingly being adopted by non-financial companies, especially for Business Activity Monitoring software (BAM). One of the most distinctive features of CEP is pattern matching. In this field, pattern matching should not be mistaken for the pattern matching in character strings, nor pattern matching from the functional programming paradigm. Rather, it specifies temporal relationships between events, such as sequences, restrictions of event occurrences, just to name a few. On the other hand, one feature which has yet to be widely adopted amongst the CEP engines is entity state and lifecycle modeling, i.e., the ability to correlate events into instances of an entity, where each entity will have di↵erent states and transitions representing its lifecycle. The objective of this work is to investigate and implement this type of functionality in Feedzai Pulse’s query language - PQL.