predictivepolicing
DESCRIPTION
Politiechef Charli Beck uit LATRANSCRIPT
![Page 1: PredictivePolicing](https://reader033.vdocuments.mx/reader033/viewer/2022060110/5560c9b0d8b42a0d088b498d/html5/thumbnails/1.jpg)
The Los Angeles Predictive Policing Experiment
Charlie Beck, Chief of Police Los Angeles Police Department
![Page 2: PredictivePolicing](https://reader033.vdocuments.mx/reader033/viewer/2022060110/5560c9b0d8b42a0d088b498d/html5/thumbnails/2.jpg)
2
![Page 3: PredictivePolicing](https://reader033.vdocuments.mx/reader033/viewer/2022060110/5560c9b0d8b42a0d088b498d/html5/thumbnails/3.jpg)
3
![Page 4: PredictivePolicing](https://reader033.vdocuments.mx/reader033/viewer/2022060110/5560c9b0d8b42a0d088b498d/html5/thumbnails/4.jpg)
4
![Page 5: PredictivePolicing](https://reader033.vdocuments.mx/reader033/viewer/2022060110/5560c9b0d8b42a0d088b498d/html5/thumbnails/5.jpg)
5
Predictive Policing Time Horizon Frequency of Statistical Review
2000 2005
2010
1995
1990 Monthly
Annually
Near
Real Time Real Time
Forecasting
![Page 6: PredictivePolicing](https://reader033.vdocuments.mx/reader033/viewer/2022060110/5560c9b0d8b42a0d088b498d/html5/thumbnails/6.jpg)
6
“A place-based approach to crime analysis that utilizes
algorithm-driven crime forecasts to inform decision making
to prevent crime”
• Not individual-based
• Not arrest-based
• Risk-based Deployment of patrol resources
• Builds on community policing
• Builds on Hot Spots
• Offers more specificity
• Sets us up for better long term strategic analysis
Predictive Policing, Our Definition
![Page 7: PredictivePolicing](https://reader033.vdocuments.mx/reader033/viewer/2022060110/5560c9b0d8b42a0d088b498d/html5/thumbnails/7.jpg)
7
How is this different? • Evidence-based rather than heuristic
• Forecasting versus Retrospective Compstat Look
• Reduces tendency to “chase the dots”
• Offers more specificity in time and space
• Should eliminate biases (arrests are not used)
• Leverages existing real-time data
• Not a massive multivariate model
• Uses Date, Time, Place and Type of crime
• Allows cops to problem solve in right place at right time
• Dosage findings should increase efficiency
![Page 8: PredictivePolicing](https://reader033.vdocuments.mx/reader033/viewer/2022060110/5560c9b0d8b42a0d088b498d/html5/thumbnails/8.jpg)
8
The Los Angeles Predictive Policing Experiment
Questions to be answered…
![Page 9: PredictivePolicing](https://reader033.vdocuments.mx/reader033/viewer/2022060110/5560c9b0d8b42a0d088b498d/html5/thumbnails/9.jpg)
9
Is it possible to predict crime?
If we predict it, can we prevent it?
If we prevent it,
how can we measure that?
Questions to be answered…
![Page 10: PredictivePolicing](https://reader033.vdocuments.mx/reader033/viewer/2022060110/5560c9b0d8b42a0d088b498d/html5/thumbnails/10.jpg)
10
Jeff Brantingham
UCLA Anthropology
George Mohler
Santa Clara Mathematics
George Tita
UCI Criminology, Law and Society
![Page 11: PredictivePolicing](https://reader033.vdocuments.mx/reader033/viewer/2022060110/5560c9b0d8b42a0d088b498d/html5/thumbnails/11.jpg)
11
Repeat Victimization
0
0.001
0.002
0.003
0.004
0.005
0.006
0 50 100 150 200 250 300 350 400
pro
ba
bil
ity
p(τ
)
days between repeat crimes (τ)
repeat crime is much more likely to happen in a short interval of time after the first event
![Page 12: PredictivePolicing](https://reader033.vdocuments.mx/reader033/viewer/2022060110/5560c9b0d8b42a0d088b498d/html5/thumbnails/12.jpg)
12
Research Design: The LAPD Experiment
• Rigorous examination of both forecasting value and of dosage
• Careful design of experiment is crucial to determine if Predictive Policing “works”
• Several designs possible, but only one – randomization – represents the “gold standard” of scientific design
![Page 13: PredictivePolicing](https://reader033.vdocuments.mx/reader033/viewer/2022060110/5560c9b0d8b42a0d088b498d/html5/thumbnails/13.jpg)
13
Daily Randomization • Foothill Division serves as both the treatment and
control area (seamless to the subjects)
• On random treatment days, the “daily mission” is determined by the Predictive Policing model; otherwise, mission is determined using existing methodology
• Experiment will run for a minimum of 6 months and evaluate whether total crimes lower on days when Predictive Policing model used
![Page 14: PredictivePolicing](https://reader033.vdocuments.mx/reader033/viewer/2022060110/5560c9b0d8b42a0d088b498d/html5/thumbnails/14.jpg)
14
![Page 15: PredictivePolicing](https://reader033.vdocuments.mx/reader033/viewer/2022060110/5560c9b0d8b42a0d088b498d/html5/thumbnails/15.jpg)
15
Foothill Patrol Area
![Page 16: PredictivePolicing](https://reader033.vdocuments.mx/reader033/viewer/2022060110/5560c9b0d8b42a0d088b498d/html5/thumbnails/16.jpg)
16
Forecasting Tool Interface v 1.0
![Page 17: PredictivePolicing](https://reader033.vdocuments.mx/reader033/viewer/2022060110/5560c9b0d8b42a0d088b498d/html5/thumbnails/17.jpg)
17
Daily Forecast Map
![Page 18: PredictivePolicing](https://reader033.vdocuments.mx/reader033/viewer/2022060110/5560c9b0d8b42a0d088b498d/html5/thumbnails/18.jpg)
18
Carl St between Borden & Chivers Ave RD 1613
Pierce St and Borden Ave RD1613
Close-up View of Mission “Boxes”
![Page 19: PredictivePolicing](https://reader033.vdocuments.mx/reader033/viewer/2022060110/5560c9b0d8b42a0d088b498d/html5/thumbnails/19.jpg)
![Page 20: PredictivePolicing](https://reader033.vdocuments.mx/reader033/viewer/2022060110/5560c9b0d8b42a0d088b498d/html5/thumbnails/20.jpg)
20
• October 19, 2011: Training on production of forecasts and methodology begins
• November 6, 2011: Three month randomized study begins
• March 2012: Evaluation results
Continued iterative process to test both the algorithm as well
as the dosage and interventions.
The Timeline
![Page 21: PredictivePolicing](https://reader033.vdocuments.mx/reader033/viewer/2022060110/5560c9b0d8b42a0d088b498d/html5/thumbnails/21.jpg)
21
Initial Observations… • Bottomline:
• How accurate are the forecasts?
• Versus random and versus state of the art
• How effective in crime fighting
• Crime numbers time series
• Dosage discussion
• Weekend vs. weekday finding
• Ease of use improvements
• 4 times per day by watch
• Roll out plans
![Page 22: PredictivePolicing](https://reader033.vdocuments.mx/reader033/viewer/2022060110/5560c9b0d8b42a0d088b498d/html5/thumbnails/22.jpg)
22
Preliminary results
• soft numbers pending completion of statistical study
• focusing on weekdays only
• accuracy of PredPol algorithm vs. best-practice
• PredPol vs. independent control group: 10.6% vs. 9.8%
• PredPol applied to control group: 11.4% vs. 9.8%
• PredPol is 8-16% more accurate compared to best practice
![Page 23: PredictivePolicing](https://reader033.vdocuments.mx/reader033/viewer/2022060110/5560c9b0d8b42a0d088b498d/html5/thumbnails/23.jpg)
23 Predictive Policing is promising in historical data analysis
![Page 24: PredictivePolicing](https://reader033.vdocuments.mx/reader033/viewer/2022060110/5560c9b0d8b42a0d088b498d/html5/thumbnails/24.jpg)
24
66
60 65 67
40 43
40
51
37 42
0
10
20
30
40
50
60
70
80
week 1 week 2 week 3 week 4 week 5
Crime Numbers (Burg, BFMV & GTA)
Test Period Weeks 1 thru 5 - 2010 v. 2011
2010 avg.= 60
2011 avg.= 50
Test Period 2010 avg.= 60 Test Period 2011 avg.= 42
2010
20
11
![Page 25: PredictivePolicing](https://reader033.vdocuments.mx/reader033/viewer/2022060110/5560c9b0d8b42a0d088b498d/html5/thumbnails/25.jpg)
25
Get Creative…
Securing Buy-In from Patrol
![Page 26: PredictivePolicing](https://reader033.vdocuments.mx/reader033/viewer/2022060110/5560c9b0d8b42a0d088b498d/html5/thumbnails/26.jpg)
26
Patrol Activity (addressing forecasted mission areas)
Test Period Weeks 1 thru 5
Avg Hours in the Box per Week = 35 hrs
18 27
72
18
42 40 47
96
51
120
0
20
40
60
80
100
120
140
1 2 3 4 5
Avg Checks per Week = 71 checks
Occupy LA
= Hours “in the Box” per Week
= Checks per Week
![Page 27: PredictivePolicing](https://reader033.vdocuments.mx/reader033/viewer/2022060110/5560c9b0d8b42a0d088b498d/html5/thumbnails/27.jpg)
28
New Version of Software… • Weekend vs. weekday finding
• Ease of use improvements
• 4 times per day by watch
• Roll out plans
![Page 28: PredictivePolicing](https://reader033.vdocuments.mx/reader033/viewer/2022060110/5560c9b0d8b42a0d088b498d/html5/thumbnails/28.jpg)
29
![Page 29: PredictivePolicing](https://reader033.vdocuments.mx/reader033/viewer/2022060110/5560c9b0d8b42a0d088b498d/html5/thumbnails/29.jpg)
30
![Page 30: PredictivePolicing](https://reader033.vdocuments.mx/reader033/viewer/2022060110/5560c9b0d8b42a0d088b498d/html5/thumbnails/30.jpg)
31
![Page 31: PredictivePolicing](https://reader033.vdocuments.mx/reader033/viewer/2022060110/5560c9b0d8b42a0d088b498d/html5/thumbnails/31.jpg)
The Los Angeles Predictive Policing Experiment
Charlie Beck, Chief of Police Los Angeles Police Department
![Page 32: PredictivePolicing](https://reader033.vdocuments.mx/reader033/viewer/2022060110/5560c9b0d8b42a0d088b498d/html5/thumbnails/32.jpg)
33
Additional Slides if Needed
![Page 33: PredictivePolicing](https://reader033.vdocuments.mx/reader033/viewer/2022060110/5560c9b0d8b42a0d088b498d/html5/thumbnails/33.jpg)
34
Foothill “dosage” evidence: Number of Unique Boxes Patrolled
mean
1SD
1SD
![Page 34: PredictivePolicing](https://reader033.vdocuments.mx/reader033/viewer/2022060110/5560c9b0d8b42a0d088b498d/html5/thumbnails/34.jpg)
35
mean
1SD
1SD
![Page 35: PredictivePolicing](https://reader033.vdocuments.mx/reader033/viewer/2022060110/5560c9b0d8b42a0d088b498d/html5/thumbnails/35.jpg)
36
mean
1SD
1SD
![Page 36: PredictivePolicing](https://reader033.vdocuments.mx/reader033/viewer/2022060110/5560c9b0d8b42a0d088b498d/html5/thumbnails/36.jpg)
41
Santa Cruz • Very basic crime analysis prior to Predictive Policing
• Updated daily, the prediction maps identify “boxes” to be patrolled “when free”
• Promising results, but limited ability to do careful evaluation
• is this because Santa Cruz is now doing crime analysis where before they were not?
• no controlled comparisons
![Page 37: PredictivePolicing](https://reader033.vdocuments.mx/reader033/viewer/2022060110/5560c9b0d8b42a0d088b498d/html5/thumbnails/37.jpg)
42
![Page 38: PredictivePolicing](https://reader033.vdocuments.mx/reader033/viewer/2022060110/5560c9b0d8b42a0d088b498d/html5/thumbnails/38.jpg)
43