美國(guó)科學(xué)家開(kāi)發(fā)新算法：可提前一周預(yù)測(cè)犯罪

Advances in artificial intelligence and machine learning have sparked interest from governments that would like to use these tools for predictive policing to deter crime. However, early efforts at crime prediction have been controversial, because they do not account for systemic biases in police enforcement and its complex relationship with crime and society.
人工智能和機(jī)器學(xué)習(xí)的進(jìn)步受到各國(guó)政府關(guān)注，他們希望利用這些工具進(jìn)行預(yù)測(cè)性用警，以遏制犯罪。然而，早期的犯罪預(yù)測(cè)工作一直存在爭(zhēng)議，沒(méi)有考慮到警察執(zhí)法中的系統(tǒng)性偏見(jiàn)及其與犯罪和社會(huì)的復(fù)雜關(guān)系。

University of Chicago data and social scientists have developed a new algorithm that forecasts crime by learning patterns in time and geographic locations from public data on violent and property crimes. It has demonstrated success at predicting future crimes one week in advance with approximately 90% accuracy.
芝加哥大學(xué)數(shù)據(jù)和社會(huì)科學(xué)家已經(jīng)開(kāi)發(fā)了一種新算法，通過(guò)從暴力和財(cái)產(chǎn)犯罪的公共數(shù)據(jù)中學(xué)習(xí)時(shí)間和地理位置模式來(lái)預(yù)測(cè)犯罪。該算法成功提前一周預(yù)測(cè)未來(lái)犯罪，準(zhǔn)確率約為90%。

The new study was published on June 30, 2022, in the journal Nature Human Behavior.
該研究論文于2022年6月30日發(fā)表在《自然——人類(lèi)行為》雜志上

The new tool was tested and validated using historical data from the City of Chicago around two broad categories of reported events: violent crimes (homicides, assaults and batteries) and property crimes (burglaries, thefts, and motor vehicle thefts).
新模型使用芝加哥市的歷史數(shù)據(jù)，圍繞暴力犯罪（殺人、毆打和人身攻擊）和財(cái)產(chǎn)犯罪（入室盜竊、偷盜和機(jī)動(dòng)車(chē)盜竊）兩大類(lèi)報(bào)告案件進(jìn)行測(cè)試和驗(yàn)證。

The new model isolates crime by looking at the time and spatial coordinates of discrete events and detecting patterns to predict future events. It divides the city into spatial tiles roughly 1,000 feet across and predicts crime within these areas instead of relying on traditional neighborhood or political boundaries, which are also subject to bias. The model performed just as well with data from seven other US cities: Atlanta, Austin, Detroit, Los Angeles, Philadelphia, Portland, and San Francisco.
該模型通過(guò)觀察離散事件的時(shí)間和空間坐標(biāo)，檢測(cè)模式以預(yù)測(cè)未來(lái)事件，從而預(yù)防犯罪。它將城市劃分為每個(gè)大約300米寬的片區(qū)，并預(yù)測(cè)這些區(qū)域內(nèi)的犯罪，而不是依賴(lài)傳統(tǒng)的鄰里或行政邊界，因?yàn)檫@些邊界也會(huì)有偏差。該模型在亞特蘭大、奧斯汀、底特律、洛杉磯、費(fèi)城、波特蘭和舊金山這七個(gè)美國(guó)城市的數(shù)據(jù)中表現(xiàn)同樣出色。

Ishanu Chattopadhyay, Assistant Professor of Medicine at UChicago and senior author of the study, is careful to note that the tool’s accuracy does not mean that it should be used to direct law enforcement, with police departments using it to swarm neighborhoods proactively to prevent crime. Instead, it should be added to a toolbox of urban policies and policing strategies to address crime.
該研究論文第一作者、芝加哥大學(xué)醫(yī)學(xué)助理教授伊山·查托帕迪亞稱(chēng)謹(jǐn)慎指出，該工具的準(zhǔn)確性并不意味著應(yīng)該將其用于指導(dǎo)執(zhí)法，讓警方主動(dòng)進(jìn)入社區(qū)預(yù)防犯罪。相反，它應(yīng)該應(yīng)用于城市政策和治安策略中，以應(yīng)對(duì)犯罪。

“We created a digital twin of urban environments. If you feed it data from happened in the past, it will tell you what’s going to happen in future. It’s not magical, there are limitations, but we validated it and it works really well,” Chattopadhyay said. “Now you can use this as a simulation tool to see what happens if crime goes up in one area of the city, or there is increased enforcement in another area. If you apply all these different variables, you can see how the systems evolves in response.”
查托帕迪亞說(shuō)：“我們模擬一個(gè)數(shù)字化的城市環(huán)境。如果你向它提供過(guò)去發(fā)生的數(shù)據(jù)，它會(huì)預(yù)測(cè)出未來(lái)會(huì)發(fā)生什么。這并不神奇，也存在一些局限性，但我們對(duì)其進(jìn)行了驗(yàn)證，并且效果非常好?，F(xiàn)在，你可以把它當(dāng)作一個(gè)模擬工具，看看如果城市某個(gè)地區(qū)的犯罪率上升，或者另一個(gè)地區(qū)的執(zhí)法力度加大，會(huì)發(fā)生什么。如果你應(yīng)用所有這些不同的變量，你可以看到系統(tǒng)是如何應(yīng)對(duì)的。”