The man who saved New York City using Data Science

The man who saved New York City using Data Science

#man #saved #York #City #Data #Science

Before i start this video i just want to thank dadaiku for partnering with me to make this video possible dadaiku is the platform for everyday ai what i love about them is they make managing data and ai so easy that non-coders can be autonomous with their data prep

Analytics encoders can only step in for fancier more interesting problems to solve but more on that at the end how does a few pins on a map end up becoming the biggest crime tracking system in the us this is a story of how comstat was born a system that pushed police departments

To become more data-driven which changed the course of law enforcement in new york city to understand the significance of this we first have to go back 40 years the bloodiest year in the history of new york new york preliminary murder rate that released today is 784 homicides we’re starting off the new

Year with at least 10 murders in the 1980s new york city was a lawless place crime had reached epidemic proportions over 2 000 murders and 600 000 serious felonies committed every year the city seemed beyond redemption but there was one person who saw it differently and if

Given the chance he believed he could fix the city his name was jack mabel Jack maple plotted every single crime onto a map each pin represented a single crime color coded and marked to represent the time of day and the different crimes that allowed him to find patterns in a way looking at raw numbers simply could not it also allowed

Him to be one step ahead of the perpetrators jack called them the charts of the future but not everyone understood them picasso’s still going at it man [ __ ] that guy his old transit cop acting like he runs his place 100 bucks says he gets can within three months yeah more like two

Hey i wish i could do arts and crafts all day too lieutenant what is this it’s what’s gonna save this city bill come on See these great pins wolf pack robberies same time same place every night this this uh pickpockets in times square i’ll follow the red pinch move to downtown on west 4th every afternoon what are these purse snatchings notice anything same tray line and at the same time too

See with jack’s map crime patterns were visualized meaning they could be predicted those yellow pins on the map they let the police know where and when they’d catch next per snatching turns out it was just one guy’s doing once he was in cuffs poof no more yellow pins on the map

With jack sharks the future the nypd realized most of the crimes in their city were actually done by a small percentage of people track them understand them and arrest them it’s time we start fighting crime not just responding to it and just like that felony crime at a

Subway dropped 30 in two years outside new york city was still a cesspool but the subways were safer than ever before in 1994 new york city got a new police commissioner bill bratton he’d seen what jack maple had accomplished with the charts of the future so he wanted him to

Do the same thing but this time for all of new york city so he made him his deputy and together they would turn the charts of the future into what we know today as com staff want every to track every single crime that happens daily every robbery every murder even a

Sneeze and i want to know about it okay nothing slips through the cracks i need to hold our own people accountable they already report them to the commission and things have been pretty quiet what quiet we’ve already had 30 murders this week last year we had 1946.

Look quiet my ass this is the murder capital of the [ __ ] world they’re ignoring cases all right police departments just won’t tell you about it you need to force them to take every single crime seriously from now on well if they’re keeping the bad news away from the commission aren’t they

Gonna do the same to you well then we gotta force it out of them how are we gonna do that sir i’ll just kindly ask them of course lots of robberies in the fifth precinct what’s going on commander a lot more heroin up there sir really no kidding where is it

It’s uh where’s it coming in who’s dealing we haven’t really how do you know that it’s the uh heroinetics that are doing these robberies well as i said before we have an active investigation oh what does that mean now tell me exactly what that means

Now you woke up in the morning you went to work what exactly did you do to catch these guys if you don’t got answers i sure as hell will find someone who does dismissed last year we recovered 800 000 from the boa robbery six months ago the four-year-old son of

A world-renowned pianist was found thanks to us now we’re working on our biggest homicide case yet that’s what we do we don’t have the resources nor the time for these low-level cases where do you live commander i don’t see how that’s clockstown one of these places

What’d you pay for your house 300 000 350. got a nice little police department there huh now wouldn’t you be on the phone with the clocks down [ __ ] cops and wouldn’t you want these crackheads arrested and if they say oh geez sir don’t you understand they’re just low-level guys

Now we’re working on a big case and we’ll be done with that big case in about i don’t know a year would that be all right if your children were stepping on crack biles on their way to school so why the [ __ ] would it be all right for your precinct i’ll get out

Police chiefs hated comstat from the meetings to the date of reporting but that forced him to take every single crime seriously because of that the crime rate in new york city plunged only a year after bretton took over the murder rate dropped 25 percent a year after that 40 percent

Comstat worked it revolutionized the department and became a symbol of police accountability things were going well for new york until people who didn’t understand the machine started running it when jack stepped down in 1996 people started using comstat as a management tool setting impossible numbers to hit

Rather than making the city safer it became a numbers game year after year the crime rate would go down but what was really behind those numbers apparently due to the immense pressure to lower the crime rate police chiefs started fudging numbers downgrading crimes or even stop reporting them

For some it was the only way not to get fired they were so focused on moving the metric it no longer solved the actual problem we have to remember that behind those numbers and statistics are real people with their own individual stories and no story can be reduced into a single number

That’s why we’ll always need the human element in a data-driven system to keep it in check comstat’s data-driven approach remains a controversial one on the one hand it almost definitely led to smarter policing and a safer city on the other hand it became justification for police abuses perhaps

The answer is that like all machines they can only be as effective as the people running it and the data that goes into it data is not a solution but a tool to be used in the hands of responsible leaders and that’s where you come in thanks for watching this video if you

Like this story there are plenty more to discover on history of data and you can even try to win the graphic novel innovators of data science which tells the stories of the 12 people who had the most significant impact on the history of data science on the website

History of data science you’ll learn how data science was used to fight the kova 19 pandemic how it was used back in the 19th century to fight cholera what a.i winter means why the dartmouth conference was so important and a lot more cool stuff and this is all done by

Dadeku which is a sponsor of this video so about dadaiku it’s a platform for everyday ai it makes designing deploying and managing ai super easy for me since i was mainly doing analytics at facebook i think their analytics application would have been extremely helpful i’m not a big fan of writing code just

For the sake of writing code because it just slows you down but with their platform instead of writing custom scripts to clean your data create pipelines and deploy your model it’s all in one easy to use platform this way you can focus on solving more interesting machine learning problems

Alright that’s it i’ll see you in the next one

Like it? Share with your friends!


What's Your Reaction?

hate hate
confused confused
fail fail
fun fun
geeky geeky
love love
lol lol
omg omg
win win


Choose A Format
Voting to make decisions or determine opinions
Formatted Text with Embeds and Visuals
Youtube and Vimeo Embeds
Soundcloud or Mixcloud Embeds
Photo or GIF
GIF format