Knife crime: assault data can help forecast fatal stabbings

fpjl2 — Fri, 12 Apr 2019 10:23:41 +0000

Knife crime data from a 12-month period could be used to help forecast the London neighbourhoods most likely to suffer a fatal stabbing the following year, according to latest research.

Helping police make custody decisions using artificial intelligence

lw355 — Mon, 26 Feb 2018 13:20:08 +0000

"It鈥檚 3am on Saturday morning. 探花直播man in front of you has been caught in possession of drugs. He has no weapons, and no record of any violent or serious crimes. Do you let the man out on police bail the next morning, or keep him locked up for two days to ensure he comes to court on Monday?鈥�

探花直播kind of scenario Dr Geoffrey Barnes is describing 鈥撀爓hether to detain a suspect in police custody or release them on bail聽鈥撀爋ccurs hundreds of thousands of times a year across the UK. 探花直播outcome of this decision could be major for the suspect, for public safety and for the police.

鈥� 探花直播police officers who make these custody decisions are highly experienced,鈥� explains Barnes. 鈥淏ut all their knowledge and policing skills can鈥檛 tell them the one thing they need to now most about the suspect 鈥� how likely is it that he or she is going to cause major harm if they are released? This is a job that really scares people 鈥� they are at the front line of risk-based decision-making.鈥�

Barnes and Professor Lawrence Sherman, who leads the Jerry Lee Centre for Experimental Criminology in the 探花直播 of Cambridge鈥檚 Institute of Criminology, have been working with police forces around the world to ask whether AI can help.聽

鈥淚magine a situation where the officer has the benefit of a hundred thousand, and more, real previous experiences of custody decisions?鈥� says Sherman. 鈥淣o one person can have that number of experiences, but a machine can.鈥�

In mid-2016, with funding from the Monument Trust, the researchers installed the world鈥檚 first AI tool for helping police make custodial decisions in Durham Constabulary.

Called the Harm Assessment Risk Tool (HART), the AI-based technology uses 104,000 histories of people previously arrested and processed in Durham custody suites over the course of five years, with a two-year follow-up for each custody decision. Using a method called 鈥渞andom forests鈥�, the model looks at vast numbers of combinations of 鈥榩redictor values鈥�, the majority of which focus on the suspect鈥檚 offending history, as well as age, gender and geographical area.聽

鈥淭hese variables are combined in thousands of different ways before a final forecasted conclusion is reached,鈥� explains Barnes. 鈥淚magine a human holding this number of variables in their head, and making all of these connections before making a decision. Our minds simply can鈥檛 do it.鈥�

探花直播aim of HART is to categorise whether in the next two years an offender is high risk (highly likely to commit a new serious offence such as murder, aggravated violence, sexual crimes or robbery); moderate risk (likely to commit a non-serious offence); or low risk (unlikely to commit any offence).聽

鈥� 探花直播need for good prediction is not just about identifying the dangerous people,鈥� explains Sherman. 鈥淚t鈥檚 also about identifying people who definitely are not dangerous. For every case of a suspect on bail who kills someone, there are tens of thousands of non-violent suspects who are locked up longer than necessary.鈥�

Durham Constabulary want to identify the 鈥榤oderate-risk鈥� group 鈥� who account for just under half of all suspects according to the statistics generated by HART. These individuals might benefit from their Checkpoint programme, which aims to tackle the root causes of offending and offer an alternative to prosecution that they hope will turn moderate risks into low risks.聽

鈥淚t鈥檚 needles and haystacks,鈥� says Sherman. 鈥淥n the one hand, the dangerous 鈥榥eedles鈥� are too rare for anyone to meet often enough to spot them on sight. On the other, the 鈥榟ay鈥� poses no threat and keeping them in custody wastes resources and may even do more harm than good.鈥� A randomised controlled trial is currently under way in Durham to test the use of Checkpoint among those forecast as moderate risk.

HART is also being refreshed with more recent data 鈥� a step that Barnes explains will be an important part of this sort of tool: 鈥淎 human decision-maker might adapt immediately to a changing context 鈥� such as a prioritisation of certain offences, like hate crime 鈥� but the same cannot necessarily be said of an algorithmic tool. This suggests the need for careful and constant scrutiny of the predictors used and for frequently refreshing the algorithm with more recent historical data.鈥�

No prediction tool can be perfect. An independent validation study of HART found an overall accuracy of around 63%. But, says Barnes, the real power of machine learning comes not from the avoidance of any error at all but from deciding which errors you most want to avoid.聽

鈥淣ot all errors are equal,鈥� says Sheena Urwin, head of criminal justice at Durham Constabulary and a graduate of the Institute of Criminology鈥檚 Police Executive Master of Studies Programme. 鈥� 探花直播worst error would be if the model forecasts low and the offender turned out high.鈥�

鈥淚n consultation with the Durham police, we built a system that is 98% accurate at avoiding this most dangerous form of error 鈥� the 鈥榝alse negative鈥� 鈥� the offender who is predicted to be relatively safe, but then goes on to commit a serious violent offence,鈥� adds Barnes. 鈥淎I is infinitely adjustable and when constructing an AI tool it鈥檚 important to weigh up the most ethically appropriate route to take.鈥�

探花直播researchers also stress that HART鈥檚 output is for guidance only, and that the ultimate decision is that of the police officer in charge.

鈥淗ART uses Durham鈥檚 data and so it鈥檚 only relevant for offences committed in the jurisdiction of Durham Constabulary. This limitation is one of the reasons why such models should be regarded as supporting human decision-makers not replacing them,鈥� explains Barnes. 鈥淭hese technologies are not, of themselves, silver bullets for law enforcement, and neither are they sinister machinations of a so-called surveillance state.鈥�

Some decisions, says Sherman, have too great an impact on society and the welfare of individuals for them to be influenced by an emerging technology.

Where AI-based tools provide great promise, however, is to use the forecasting of offenders鈥� risk level for effective 鈥榯riage鈥�, as Sherman describes: 鈥� 探花直播police service is under pressure to do more with less, to target resources more efficiently, and to keep the public safe.聽

鈥� 探花直播tool helps identify the few 鈥榥eedles in the haystack鈥� who pose a major danger to the community, and whose release should be subject to additional layers of review. At the same time, better triaging can lead to the right offenders receiving release decisions that benefit both them and society.鈥�

Inset image: read more about our AI research in the 探花直播's research magazine;聽download聽a聽pdf;聽view聽on聽Issuu.

Police at the 鈥渇ront line鈥� of difficult risk-based judgements are trialling an AI system trained by 探花直播 of Cambridge criminologists to give guidance using the outcomes of five years of criminal histories.

探花直播tool helps identify the few 鈥榥eedles in the haystack鈥� who pose a major danger to the community, and whose release should be subject to additional layers of review

Lawrence Sherman

Rene B枚hmer on Unsplash

探花直播text in this work is licensed under a Creative Commons Attribution 4.0 International License. For image use please see separate credits above.

Yes

探花直播 of Cambridge - Jerry Lee Centre for Experimental Criminology

Knife crime: assault data can help forecast fatal stabbings

Helping police make custody decisions using artificial intelligence