The document discusses using data-driven approaches to make security decisions. Specifically, it discusses using data collected about past incidents to predict future risks and enable decision makers to take appropriate actions. An example tool called Insights is mentioned that uses real-time data to provide localized risk assessments and early warnings about potential crimes. The key is having large volumes of detailed data from various sources that can be analyzed using technologies like machine learning to help mitigate security risks.
2. Security decisions are often made based on experience, local understanding and
standard mitigations
• Previous experience
• Country and local knowledge
• Standard mitigations and
actions
3. • What can happen?
• When and how will it happen?
• What should I do and what
impact will it have?
Being data-driven in security is the ability collect, analyze and act on data rather than
individual experience
4. Being data-driven is about enabling us to make the
right decision, at the right moment, to mitigate or
reduce the impact of an incident
5. One example of this is Insights, a localized risk assessment tool
6.
7. Transform incident records to
predictive risk assessments
Example use of real-time risk assessment: Early warning for burglaries by predicting
on crime waves
8. Real-time operation example: Utilizing early indicators real-time to reduce or avoid
the impact of an ongoing incident
9. Statistics Heat map Specific countermeasures Business caseRealtime risk
AI technology is being commoditized and the key asset is high volumes
of detailed data
Incident volumes Location Time Modus operandi Business impact
10. Where do you start?
Use the data you have. It is likely not tailored for machine learning but
it will get you started.
1
Create a vision on what you want to achieve. Make changes to your
processes to capture the needed data.
2
Partner with others – combining your data with others can be
beneficial to all parties
3