Her Majesty’s Inspectorate of Police and Fire and Rescue Services (HMICFRS) to regularly inspect and monitor the police. leather (Evaluation of police effectiveness, efficiency, and legitimacy).
If a serious issue is discovered and the response is deemed insufficient, police will escalate to enhanced surveillance (also known as engage). However, by this stage, services to the public may already have been affected.
HMICFRS Accelerated Capability Environmental (ace) to see if there is a way to create an early warning prediction tool that can estimate . leather Evaluation grade before inspection. This results in HMICFRSThe inspection program will prioritize more urgent visits to reported units, with the aim of remediating problems early and making communities safer.
Apply machine learning to crime data
The decision was made to focus on one of them. leather Evaluation Question: How well does the military investigate crimes? Working with supplier The London Data Company, a proof of concept for the machine learning algorithm was put together in just his eight weeks.
This used publicly available data on levels of crime and crime outcomes from sources such as 999 calls, the Home Office and the Office for National Statistics.I was able to predict correctly leather Force grades were set in approximately 60% of cases, and were graded one grade higher or lower in approximately 90% of cases.
Jackie Hayes HMICFRS Insight Portfolio Director said: “Our testing process looks at vast amounts of data from our forces, and this tool, broadly speaking, comes to very similar conclusions.
“We are currently exploring what more we can do with the data we collect and what else we can do. leather I have a question that can expand on this. ”
HMICFRShas ambitions to become more data-driven and is currently working with the London Data Company on how to introduce initial demonstrator tools into live systems and overall inspection processes over the next 18 months. is.
Regarding the next potential applications, Hayes said: “Fire and rescue is also on the list, but it's a very long list because we want to do a lot with it.”
“You can't replace testing teams with artificial intelligence, but you can certainly think about what this means for testing methods. I think this will have an impact on that.”
Furthermore, she added: “This tells us where the problem might be, but it doesn't tell us why it's a problem. So we follow up on our inspection activities to see if it's correct and what's causing the problem. We want to see what it is,” and what the powers that be are doing about it.
“We want to make our communities safer, so we're not just pointing out the problems, we're trying to support the forces that make them better.”
