![Page 1: Crime Risk Forecasting: Near Repeat Pattern Analysis & Load Forecasting](https://reader033.vdocument.in/reader033/viewer/2022050920/54b3a88b4a7959c36d8b46ca/html5/thumbnails/1.jpg)
340 N 12th St, Suite 402Philadelphia, PA 19107
www.azavea.com/hunchlab
Crime Risk Forecasting
Near Repeat Pattern Analysis and Load Forecasting
![Page 2: Crime Risk Forecasting: Near Repeat Pattern Analysis & Load Forecasting](https://reader033.vdocument.in/reader033/viewer/2022050920/54b3a88b4a7959c36d8b46ca/html5/thumbnails/2.jpg)
About Us
Robert CheethamPresident & [email protected]
Jeremy HeffnerHunchLab Product [email protected]
![Page 3: Crime Risk Forecasting: Near Repeat Pattern Analysis & Load Forecasting](https://reader033.vdocument.in/reader033/viewer/2022050920/54b3a88b4a7959c36d8b46ca/html5/thumbnails/3.jpg)
Agenda
• Company Background• HunchLab
– Risk Forecasting• Near Repeat Pattern Analysis• Load Forecasting
– Future Research Topics
• Q&A
![Page 4: Crime Risk Forecasting: Near Repeat Pattern Analysis & Load Forecasting](https://reader033.vdocument.in/reader033/viewer/2022050920/54b3a88b4a7959c36d8b46ca/html5/thumbnails/4.jpg)
About Azavea
• Founded in 2000
• 25 people
• Based in Philadelphia
– Boston satellite office
• Geospatial + web + mobile
– Software development
– Spatial analysis services
![Page 5: Crime Risk Forecasting: Near Repeat Pattern Analysis & Load Forecasting](https://reader033.vdocument.in/reader033/viewer/2022050920/54b3a88b4a7959c36d8b46ca/html5/thumbnails/5.jpg)
Clients & Industries
• Public Safety• Municipal Services• Public Health• Human Services• Culture • Elections & Politics• Land Conservation• Economic Development
![Page 6: Crime Risk Forecasting: Near Repeat Pattern Analysis & Load Forecasting](https://reader033.vdocument.in/reader033/viewer/2022050920/54b3a88b4a7959c36d8b46ca/html5/thumbnails/6.jpg)
Azavea & Governments
![Page 7: Crime Risk Forecasting: Near Repeat Pattern Analysis & Load Forecasting](https://reader033.vdocument.in/reader033/viewer/2022050920/54b3a88b4a7959c36d8b46ca/html5/thumbnails/7.jpg)
HunchLab
![Page 8: Crime Risk Forecasting: Near Repeat Pattern Analysis & Load Forecasting](https://reader033.vdocument.in/reader033/viewer/2022050920/54b3a88b4a7959c36d8b46ca/html5/thumbnails/8.jpg)
web-based crime analysis, early warning, and risk forecasting
![Page 9: Crime Risk Forecasting: Near Repeat Pattern Analysis & Load Forecasting](https://reader033.vdocument.in/reader033/viewer/2022050920/54b3a88b4a7959c36d8b46ca/html5/thumbnails/9.jpg)
Crime Analysis
– Mapping (spatial / temporal densities)
– Trending
– Intelligence Dashboard
Early Warning
– Statistical & Threshold-based Hunches (data mining)
– Alerting
Risk Forecasting
– Near Repeat Pattern
– Load Forecasting
![Page 10: Crime Risk Forecasting: Near Repeat Pattern Analysis & Load Forecasting](https://reader033.vdocument.in/reader033/viewer/2022050920/54b3a88b4a7959c36d8b46ca/html5/thumbnails/10.jpg)
![Page 11: Crime Risk Forecasting: Near Repeat Pattern Analysis & Load Forecasting](https://reader033.vdocument.in/reader033/viewer/2022050920/54b3a88b4a7959c36d8b46ca/html5/thumbnails/11.jpg)
Near Repeat Pattern Analysis
![Page 12: Crime Risk Forecasting: Near Repeat Pattern Analysis & Load Forecasting](https://reader033.vdocument.in/reader033/viewer/2022050920/54b3a88b4a7959c36d8b46ca/html5/thumbnails/12.jpg)
Contagious Crime?
• Near repeat pattern analysis • “If one burglary occurs, how does the risk change nearby?”
![Page 13: Crime Risk Forecasting: Near Repeat Pattern Analysis & Load Forecasting](https://reader033.vdocument.in/reader033/viewer/2022050920/54b3a88b4a7959c36d8b46ca/html5/thumbnails/13.jpg)
What Do We Mean By Near Repeat?
• Repeat victimization– Incident at the same location at a later time (likely
related)
• Near repeat victimization– Incident at a nearby location at a later time (likely
related)
• Incident A (place, time) --> Incident B (place, time)
![Page 14: Crime Risk Forecasting: Near Repeat Pattern Analysis & Load Forecasting](https://reader033.vdocument.in/reader033/viewer/2022050920/54b3a88b4a7959c36d8b46ca/html5/thumbnails/14.jpg)
Near Repeat Pattern Analysis
• The goal:– Quantify short term risk due to near-repeat victimization
• “If one burglary occurs, how does the risk of burglary for the neighbors change?”
• What we know:– Incident A (place, time) --> Incident B (place, time)
• Distance between A and B• Timeframe between A and B
• What we need to know:– What distances/timeframes are not simply random?
![Page 15: Crime Risk Forecasting: Near Repeat Pattern Analysis & Load Forecasting](https://reader033.vdocument.in/reader033/viewer/2022050920/54b3a88b4a7959c36d8b46ca/html5/thumbnails/15.jpg)
Near Repeat Pattern Analysis
• The process– Observe the pattern in historic data– Simulate the pattern in randomized historic data– Compare the observed pattern to the simulated patterns– Apply the non-random pattern to new incidents
• An example– 180 days of burglaries in Division 6 of Philadelphia
![Page 16: Crime Risk Forecasting: Near Repeat Pattern Analysis & Load Forecasting](https://reader033.vdocument.in/reader033/viewer/2022050920/54b3a88b4a7959c36d8b46ca/html5/thumbnails/16.jpg)
Near Repeat Pattern Analysis
![Page 17: Crime Risk Forecasting: Near Repeat Pattern Analysis & Load Forecasting](https://reader033.vdocument.in/reader033/viewer/2022050920/54b3a88b4a7959c36d8b46ca/html5/thumbnails/17.jpg)
Near Repeat Pattern Analysis
![Page 18: Crime Risk Forecasting: Near Repeat Pattern Analysis & Load Forecasting](https://reader033.vdocument.in/reader033/viewer/2022050920/54b3a88b4a7959c36d8b46ca/html5/thumbnails/18.jpg)
Near Repeat Pattern Analysis
![Page 19: Crime Risk Forecasting: Near Repeat Pattern Analysis & Load Forecasting](https://reader033.vdocument.in/reader033/viewer/2022050920/54b3a88b4a7959c36d8b46ca/html5/thumbnails/19.jpg)
Near Repeat Pattern Analysis
![Page 20: Crime Risk Forecasting: Near Repeat Pattern Analysis & Load Forecasting](https://reader033.vdocument.in/reader033/viewer/2022050920/54b3a88b4a7959c36d8b46ca/html5/thumbnails/20.jpg)
Near Repeat Pattern Analysis
• How can you test your own data?– Near Repeat Calculator
• http://www.temple.edu/cj/misc/nr/
• Papers– Near-Repeat Patterns in Philadelphia Shootings (2008)
• One city block & two weeks after one shooting– 33% increase in likelihood of a second event
Jerry RatcliffeTemple University
![Page 21: Crime Risk Forecasting: Near Repeat Pattern Analysis & Load Forecasting](https://reader033.vdocument.in/reader033/viewer/2022050920/54b3a88b4a7959c36d8b46ca/html5/thumbnails/21.jpg)
Demo
![Page 22: Crime Risk Forecasting: Near Repeat Pattern Analysis & Load Forecasting](https://reader033.vdocument.in/reader033/viewer/2022050920/54b3a88b4a7959c36d8b46ca/html5/thumbnails/22.jpg)
Load Forecasting
![Page 23: Crime Risk Forecasting: Near Repeat Pattern Analysis & Load Forecasting](https://reader033.vdocument.in/reader033/viewer/2022050920/54b3a88b4a7959c36d8b46ca/html5/thumbnails/23.jpg)
Improving CompStat
• Load forecasting• “Given the time of year, day of week, time of day and
general trend, what counts of crimes should I expect?”
![Page 24: Crime Risk Forecasting: Near Repeat Pattern Analysis & Load Forecasting](https://reader033.vdocument.in/reader033/viewer/2022050920/54b3a88b4a7959c36d8b46ca/html5/thumbnails/24.jpg)
What Do We Mean By Load Forecasting?
• Load forecasting• Generating aggregate crime counts for a future timeframe
using cyclical time series analysis
Measure cyclical patterns
Identify non-cyclical trend
Forecast expected count
+
bit.ly/gorrcrimeforecastingpaper
![Page 25: Crime Risk Forecasting: Near Repeat Pattern Analysis & Load Forecasting](https://reader033.vdocument.in/reader033/viewer/2022050920/54b3a88b4a7959c36d8b46ca/html5/thumbnails/25.jpg)
Load Forecasting
• Measure cyclical patterns• Take historic incidents (for example: last five years)• Generate multiplicative seasonal indices
– For each time cycle:» time of year» day of week» time of day
– Count incidents within each time unit (for example: Monday)– Calculate average per time unit if incidents were evenly
distributed– Divide counts within each time unit by the calculated average
to generate multiplicative indices» Index ~ 1 means at the average» Index > 1 means above average» Index < 1 means below average
![Page 26: Crime Risk Forecasting: Near Repeat Pattern Analysis & Load Forecasting](https://reader033.vdocument.in/reader033/viewer/2022050920/54b3a88b4a7959c36d8b46ca/html5/thumbnails/26.jpg)
Load Forecasting
![Page 27: Crime Risk Forecasting: Near Repeat Pattern Analysis & Load Forecasting](https://reader033.vdocument.in/reader033/viewer/2022050920/54b3a88b4a7959c36d8b46ca/html5/thumbnails/27.jpg)
Load Forecasting
![Page 28: Crime Risk Forecasting: Near Repeat Pattern Analysis & Load Forecasting](https://reader033.vdocument.in/reader033/viewer/2022050920/54b3a88b4a7959c36d8b46ca/html5/thumbnails/28.jpg)
Load Forecasting
![Page 29: Crime Risk Forecasting: Near Repeat Pattern Analysis & Load Forecasting](https://reader033.vdocument.in/reader033/viewer/2022050920/54b3a88b4a7959c36d8b46ca/html5/thumbnails/29.jpg)
Load Forecasting
![Page 30: Crime Risk Forecasting: Near Repeat Pattern Analysis & Load Forecasting](https://reader033.vdocument.in/reader033/viewer/2022050920/54b3a88b4a7959c36d8b46ca/html5/thumbnails/30.jpg)
Load Forecasting
• Identify non-cyclical trend• Take recent daily counts (for example: last year daily
counts)• Remove cyclical trends by dividing by indices
• Run a trending function on the new counts– Simple average
» Last X Days
– Smoothing function» Exponential smoothing» Holt’s linear exponential smoothing
![Page 31: Crime Risk Forecasting: Near Repeat Pattern Analysis & Load Forecasting](https://reader033.vdocument.in/reader033/viewer/2022050920/54b3a88b4a7959c36d8b46ca/html5/thumbnails/31.jpg)
Load Forecasting
• Forecast expected count• Project trend into future timeframe
– Always flat» Simple average» Exponential smoothing
– Linear trend» Holt’s linear exponential smoothing
• Multiple by seasonal indices to reseasonalize the data
![Page 32: Crime Risk Forecasting: Near Repeat Pattern Analysis & Load Forecasting](https://reader033.vdocument.in/reader033/viewer/2022050920/54b3a88b4a7959c36d8b46ca/html5/thumbnails/32.jpg)
Load Forecasting
Measure cyclical patterns
Identify non-cyclical trend
Forecast expected count
+
bit.ly/gorrcrimeforecastingpaper
![Page 33: Crime Risk Forecasting: Near Repeat Pattern Analysis & Load Forecasting](https://reader033.vdocument.in/reader033/viewer/2022050920/54b3a88b4a7959c36d8b46ca/html5/thumbnails/33.jpg)
How Do We Know It’s Accurate?
• Testing• Generated forecasting packages (examples)
– Commonly Used» Average of last 30 days» Average of last 365 days» Last year’s count for the same time period
– Advanced Combinations» Different cyclical indices (example: day of year vs. month of year)» Different levels of geographic aggregation for indices» Different trending functions
• Scoring methodologies (examples)– Mean absolute percent error (with some enhancements)– Mean percent error– Mean squared error
• Run thousands of forecasts through testing framework• Choose the right technique in the right situation
![Page 34: Crime Risk Forecasting: Near Repeat Pattern Analysis & Load Forecasting](https://reader033.vdocument.in/reader033/viewer/2022050920/54b3a88b4a7959c36d8b46ca/html5/thumbnails/34.jpg)
Demo
![Page 35: Crime Risk Forecasting: Near Repeat Pattern Analysis & Load Forecasting](https://reader033.vdocument.in/reader033/viewer/2022050920/54b3a88b4a7959c36d8b46ca/html5/thumbnails/35.jpg)
Research Topics
![Page 36: Crime Risk Forecasting: Near Repeat Pattern Analysis & Load Forecasting](https://reader033.vdocument.in/reader033/viewer/2022050920/54b3a88b4a7959c36d8b46ca/html5/thumbnails/36.jpg)
Research Topics
• Analysis– Real-time Functionality
• Consume real-time data streams• Conduct ongoing, automated analysis• Push real-time alerts
• Risk Forecasting– Load forecasting enhancements
• Machine learning-based model selection• Weather and special events
– Combining short and long term risk forecasts• NIJ project with Jerry Ratcliffe & Ralph Taylor• Neighborhood composition modeling using ACS data
– Risk Terrain Modeling
![Page 37: Crime Risk Forecasting: Near Repeat Pattern Analysis & Load Forecasting](https://reader033.vdocument.in/reader033/viewer/2022050920/54b3a88b4a7959c36d8b46ca/html5/thumbnails/37.jpg)
Research Topics
• Current Implementation Funding– Local Byrne Memorial JAG solicitation due July 21, 2011
• http://www.ojp.usdoj.gov/BJA/grant/jag.html
• Research Funding
![Page 38: Crime Risk Forecasting: Near Repeat Pattern Analysis & Load Forecasting](https://reader033.vdocument.in/reader033/viewer/2022050920/54b3a88b4a7959c36d8b46ca/html5/thumbnails/38.jpg)
Q&A
![Page 39: Crime Risk Forecasting: Near Repeat Pattern Analysis & Load Forecasting](https://reader033.vdocument.in/reader033/viewer/2022050920/54b3a88b4a7959c36d8b46ca/html5/thumbnails/39.jpg)
Contact Us
Robert CheethamPresident & [email protected]
Jeremy HeffnerHunchLab Product [email protected]