A mature rules engine must show the analyst what threat a rule is addressing, why the rule fired, and what the focus of the rule represents. Each rule definition must be articulable in the business context in which it operates. If ground truth data is available it must be displayed with respect to the rule finding. Analysts must validate surfaced rule findings in a way that minimally impacts their analysis while maximizing the feedback loop used by both an automated machine learning model, rule hardening or refinement, and future research.
Several modern technologies exist to get started with machine learning.
Entity Resolution is the disambiguation of data representing real world entities. The task of reducing and resolving identities can be overwhelming considering the volume of data provided in the era of Big Data. Simplifying this task into many subtasks greatly increases the likelyhood of success. There are many stages to this practice of resolving entities.