Predictive Policing: A Leap Forward in Combating Crime
In the 2002 movie Minority Report, director Steven Spielberg imagined a future where psychics could predict crimes before they happened. Twenty years on, scientists and law enforcement agencies are using data mining and machine learning to mimic those psychics. Predictive policing has shown promise in reducing crime rates, with some successes in the US and UK. However, the lack of data in under-resourced contexts like South Africa has hindered its application. Researchers have fine-tuned the Naȯve Bayes algorithm and combined it with Recursive Feature Elimination to improve crime prediction rates. This finding holds promise for applying predictive policing in South Africa, where crime levels are high and the police force seems ill-equipped to curb them.
Key Takeaways:
- Predictive policing has shown promise in reducing crime rates, with a 47% reduction in gun incidents in a US police department and a double-digit reduction in robberies, burglaries, and thefts from motor vehicles in Manchester police in the UK.
- The researchers fine-tuned the Naȯve Bayes algorithm by combining it with Recursive Feature Elimination, which selects more significant features in a dataset and removes weaker ones.
- The enhanced Naȯve Bayes achieved a 30% improvement in crime prediction rates compared to the original Naȯve Bayes.
- The researchers applied their finessed algorithm to a popular experimental dataset extracted from the Chicago Police Department's CLEAR system, which has been used to predict and reduce crime in that American city.
- The lack of data in under-resourced contexts like South Africa has hindered the application of predictive policing, but the researchers are working on a small case study in Bellville using South African Police Service data.
- Predictive algorithms and policing are not without their flaws, and fears that these algorithms may simply reinforce racial biases have been raised.
Statistics:
- 20 years since the movie Minority Report was released, with predictive policing becoming increasingly sophisticated.
- 47% reduction in gun incidents in a US police department.
- Double-digit reduction in robberies, burglaries, and thefts from motor vehicles in Manchester police in the UK.
- 30% improvement in crime prediction rates by the enhanced Naȯve Bayes algorithm.
- 20km from Cape Town's central business district, where the researchers are conducting a small case study in Bellville using South African Police Service data.
Sources:
- In: (https://www.goodreads.com/book/show/581125.The_Minority_Report) short story by Philip K Dick.
- Science: (https://www.science.org/content/article/can-predictive-policing-prevent-crime-it-happens)
- Taylor Francis: (https://www.taylorfrancis.com/chapters/edit/10.4324/9780203118214-13/crime-pattern-theory-paul-brantingham-patricia-brantingham)
- Tandfonline: (https://www.tandfonline.com/doi/full/10.1080/01900692.2019.1575664)
- Ironsides Group: (https://www.ironsidegroup.com/case-study/predictive-policing-success-manchester-police-department/)
- Springer: (https://link.springer.com/content/pdf/10.1007/978-3-642-40994-3_33.pdf)
- IGI Global: (https://www.igi-global.com/book/data-mining-trends-applications-criminal/146986)
- ACM: (https://dl.acm.org/doi/fullHtml/10.1145/3488933.3488973)
- Kaggle: (https://www.kaggle.com/datasets/slwessels/crime-statistics-for-south-africa)
- Daily Maverick: (https://www.dailymaverick.co.za/article/2022-06-03-crime-crisis-continues-in-first-quarter-of-2022-with-women-and-children-worst-affected)
- Africa Check: (https://africacheck.org/infofinder/explore-facts/how-many-people-does-one-police-officer-serve-south-africa)
- The Conversation: (https://theconversation.com/profiles/omowunmi-isafiade-1357917)