Crime Risk Forecasting and Predictive Analytics - Esri UC
Download as PPT, PDF3 likes1,424 views
The document discusses advancements in risk forecasting and predictive analytics within the Philadelphia Police Department, emphasizing the use of spatial analysis and data mining techniques for crime mapping and early warning systems. It details the concept of near repeat pattern analysis to quantify short-term crime risks and describes cyclical forecasting methods to predict future crime counts based on historical data. The document also outlines ongoing research topics and collaborations aimed at enhancing real-time analysis and leveraging machine learning for improved forecasting.
![Risk Forecasting & Predictive Analytics HunchLab Research Update Robert Cheetham [email_address] @rcheetham Michael Urciouli [email_address]](https://image.slidesharecdn.com/20110714esriuchunchlabforecasting-110716235529-phpapp02/85/Crime-Risk-Forecasting-and-Predictive-Analytics-Esri-UC-1-320.jpg)
![Questions Michael Urciouli [email_address] @rcheetham](https://image.slidesharecdn.com/20110714esriuchunchlabforecasting-110716235529-phpapp02/85/Crime-Risk-Forecasting-and-Predictive-Analytics-Esri-UC-49-320.jpg)
Crime Risk Forecasting and Predictive Analytics - Esri UC
- 1. Risk Forecasting & Predictive Analytics HunchLab Research Update Robert Cheetham [email_address] @rcheetham Michael Urciouli [email_address]
- 2. Philadelphia Police Department Crime Analysis Unit founded in 1997 Desktop ArcGIS Analysis to support investigations Support COMPSTAT process Web-based crime analysis 3 staff
- 3. Who is Azavea? 25 people - software engineers - spatial analysts - project managers Spatial Analysis Web & Mobile High Performance Geoprocessing User Experience R&D
- 4. 10% Research Program Pro Bono Program Time-to-Give-Back Program Employee-focused Culture Projects with Social Value
- 5. web-based crime analysis, early warning, and risk forecasting
- 6. 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
- 7. Dashboard
- 8. Space + Time
- 9. Space + Time
- 10. Animation
- 11. Early Warning & Notification
- 13. Risk Forecasting
- 14. Crime Analysis Mapping (spatial / temporal densities) Trending Intelligence Dashboard Early Warning Statistical & Threshold-based Hunches (data mining) Alerting Risk Forecasting Near Repeat Pattern Cyclical Forecasting
- 17. Contagious Crime? Near repeat pattern analysis “ If one burglary occurs, how does the risk change nearby?”
- 18. What Do We Mean By Near Repeat? Repeat victimization Incident at the same location at a later time Near repeat victimization Incident at a nearby location at a later time Incident A (place, time) --> Incident B (place, time)
- 19. 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?
- 20. 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
- 21. Near Repeat Pattern Analysis An example 180 days of burglaries Philadelphia Division 6
- 22. Near Repeat Pattern Analysis An example 180 days of burglaries Philadelphia Division 6
- 23. Near Repeat Pattern Analysis An example 180 days of burglaries Philadelphia Division 6
- 24. Near Repeat Pattern Analysis
- 25. Online Version
- 26. 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 Ratcliffe Temple University
- 28. Improving CompStat “ Given the time of year, day of week, time of day and general trend, what counts of crimes should I expect?”
- 29. What Do We Mean By Load Forecasting? Generating aggregate crime counts for a future timeframe using cyclical time series analysis bit.ly/gorrcrimeforecastingpaper Measure cyclical patterns Identify non-cyclical trend Forecast expected count +
- 30. Cyclical 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
- 35. Cyclical 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
- 36. Cyclical 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
- 37. Cyclical Forecasting bit.ly/gorrcrimeforecastingpaper Measure cyclical patterns Identify non-cyclical trend Forecast expected count +
- 42. How Do We Know It Works? 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
- 43. Ongoing R&D
- 44. 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
- 45. Research Topics Current Implementation Funding Local Byrne Memorial JAG solicitation due July 21, 2011 Research Funding
- 46. Acknowledgements National Science Foundation – SBIR Program Grant #IIP-0637589 Grant #IIP-0750507 Philadelphia Police Department Jerry Ratcliffe, Temple University Tony Smith, University of Pennsylvania Peirce County, WA
- 47. Upcoming Conferences Crime Mapping Research Conference, October IACP, October
- 48. Many Thanks! © Photo used with permission from Alphafish , via Flickr.com
- 49. Questions Michael Urciouli [email_address] @rcheetham