Sign in Subscribe
Police officers

Predictive Policing Holds Promise for Reducing Crime in Under-Resourced Contexts

Strategic Insights

— 2 min read

South Africa is plagued by high crime rates and an ineffective police force. Predictive policing, which uses data mining and machine learning algorithms to predict and prevent crimes, may be a valuable tool in addressing these issues. In the past decade, predictive policing has enjoyed some successes in reducing crime rates in the US and UK. By automating the process of analyzing crime data and detecting patterns, scientists and law enforcement agencies can identify high-risk areas and individuals before they commit crimes. However, the success of predictive policing also depends on the availability of high-quality data, which has been a challenge in South Africa.

Key Takeaways:

  • Predictive policing has been successful in reducing crime rates in the US and UK by automating the analysis of crime data and detecting patterns.
  • The use of machine learning algorithms and data mining techniques can improve the accuracy of crime predictions and reduce the number of violent crimes.
  • The Naïve Bayes algorithm, a popular supervised machine learning algorithm, can be fine-tuned and improved for crime prediction by marrying it with other algorithms, such as Recursive Feature Elimination.
  • The researchers found that the finessed Naïve Bayes algorithm improved on the predictions of the Naïve Bayes by about 30%, and could either match or improve on the predictions of other algorithms.
  • The lack of high-quality data is a major challenge in applying predictive policing in South Africa, despite the potential benefits of reducing crime rates and improving public confidence in the police.
  • The researchers are currently running a small case study in Bellville, a suburb near Cape Town, using the South African Police Service data for predictive policing.
  • Predictive policing is not without its flaws, including the potential for reinforcing racial biases, and its effectiveness will depend on the continued technological improvement and availability of high-quality data.

Statistics:

  • The gun incidents at a US police department were reduced by 47% over New Year's Eve.
  • Manchester police in the UK were able to predict and reduce robberies, burglaries, and thefts from motor vehicles by double digits in the first 10 weeks of rolling out predictive measures.
  • The researchers found that the finessed Naïve Bayes algorithm improved on the predictions of the Naïve Bayes by about 30%.
  • The dataset used in the research has 12,280 unique crime incidents.
  • South Africa has one of the highest crime rates in the world, with a homicide rate of 34.4 per 100,000 people in 2020.

Sources:

  • "Artificial intelligence is used for predictive policing in the US and UK – South Africa should embrace it, too" by Omowunmi Isafiade, Senior Lecturer in Computer Science, University of the Western Cape, The Conversation -- Africa.
  • Minority Report (2002) movie.
  • The Naïve Bayes algorithm and Recursive Feature Elimination.
  • Chicago Police Department's CLEAR (Citizen Law Enforcement Analysis and Reporting) system.
  • South African Police Service data.

Read more