Hands-On Artificial Intelligence for Cybersecurity

FDPS – the best of both worlds

Expert-driven and data-driven predictive models can, therefore, be combined in an FDPS in order to exploit the benefits of both approaches to improve the accuracy of the forecasts by reducing both false negatives and false positives.

The rules-based models usually reduce the number of false negatives, though this is at the cost of an increase in false positives; in combination with data-driven models, it is possible to improve forecasts by reducing false positives.

Furthermore, as we have seen, data-driven models allow operators' feedback to be integrated with other big data sources, thus contributing to dynamically updating the FDPS.

The FDPS automated maintenance and fine-tuning activities require the implementation ...