Predictive Analytic Services
Predictive analytics, aka machine learning or artificial intelligence, is using statistical models to predict the probability of a future event occurring, or forecast a numeric value that is likely to occur in the future. Examples are predicting the number of crimes likely to occur in an area in the future or predicting the probability an individual is going to be involved in gun violence in the next year.
These are not like predictions in Minority Report, they do not call for preemptively arresting people before they have committed a crime, but uses different proactive tactics to prevent crime before it occurs.
CRIME De-Coder has extensive experience in predictive analytics, has published several papers and won multiple prediction competitions (NIJ recidivism forecasting, NASA Algae Bloom).
Spatial Based Forecasts
CRIME De-Coder has published several models to forecast the most concentrated areas of crime (Wheeler & Reuter, 2021; Wheeler & Steenbeek, 2021; Yoo & Wheeler, 2019). Below is an example million dollar hotspot in Dallas using the methodology CRIME De-Coder developed (Wheeler & Reuter (2021):
Person Based Predictions
Examples of person based forecasting are chronic offender systems (Wheeler et al., 2019), and personal risk assessments for parole or bail (Circo & Wheeler, 2022). Chronic offender systems are used by police departments and prosecutors offices to identify individuals to target specialized services. Personal risk assessments are commonly used to assign individuals certain levels of parole supervision, or identify individuals of low risk to release on their own recognizance.
Time Series Forecasting
Time series forecasting is the use of historical data to forecast the number of events likely to occur in the future. CRIME De-Coder has developed methodology to forecast rare crime data (Wheeler & Kovandzic, 2018; Yim et al., 2020), and actively monitor crime patterns to identify spikes that may demand police response (Wheeler, 2016).
Fair Application of Predictive Analytics
One critique of predictive analytics is that machine learning is racist. This is misleading – predictive policing is a method to identify areas or people that can benefit the greatest from specific interventions. CRIME De-Coder has developed methodology to make predictive analytics more fair and racially equitable (Circo & Wheeler, 2022; Wheeler, 2020).
A small number of places or people cause the most crime. Targeting resources to those highest risk people and places, in the most fair way possible, is the best approach to tackle difficult crime and violence problems in communities.
If you are interested in predictive analytics, contact the CRIME De-Coder today for a free consultation to discuss your agencies needs.
References
- Circo, G.M., & Wheeler, A.P. (2022). An Open Source Replication of a Winning Recidivism Prediction Model. International Journal of Offender Therapy and Comparative Criminology, Online First.
- Wheeler, A.P. (2016). Tables and graphs for monitoring temporal crime trends: Translating theory into practical crime analysis advice. International Journal of Police Science & Management, 18(3), 159-172.
- Wheeler, A.P. (2020). Allocating police resources while limiting racial inequality. Justice Quarterly, 37(5), 842-868.
- Wheeler, A.P., & Kovandzic, T.V. (2018). Monitoring volatile homicide trends across US cities. Homicide Studies, 22(2), 119-144.
- Wheeler, A.P., & Reuter, S. (2021). Redrawing Hot Spots of Crime in Dallas, Texas. Police Quarterly, 24(2), 159-184.
- Wheeler, A.P., & Steenbeek, W. (2021). Mapping the risk terrain for crime using machine learning. Journal of Quantitative Criminology, 37, 445-480.
- Wheeler, A.P., Worden, R.E., & Silver, J.R. (2019). The accuracy of the violent offender identification directive tool to predict future gun violence. Criminal Justice and Behavior, 46(5), 770-788.
- Yim, H.N., Riddell, J.R., & Wheeler, A.P. (2020). Is the recent increase in national homicide abnormal? Testing the application of fan charts in monitoring national homicide trends over time. Journal of Criminal Justice, 66, 101656.
- Yoo, Y., & Wheeler, A.P. (2019). Using risk terrain modeling to predict homeless related crime in Los Angeles, California. Applied Geography, 109, 102039.