How Delhi Police went from ‘Reel’ to ‘Real’ with Predictive Policing

In April 2054, Washington, DC’s PreCrime police stops murderers before they act, reducing the murder rate to zero. Murders are predicted using three mutated humans, called “Precogs”, who “previsualize” crimes by receiving visions of the future…

This is the opening scene(wikipedia) of 2002 American neo-noir science fiction film called Minority Report which was quite a hit for its fictional plot. However, the Crime departments all over the world have been able to transform this fiction into reality much before 2054, thanks to Big data and Predictive Analysis.

This trend, popularly known as ‘Predictive Policing’ has already become popular in countries such as the US, UK, and China, where the authorities are not just using data to understand past criminal activities, but are also trying to predict the future crime pattern. According to Wikipedia, Predictive Policing uses data on the times, locations and nature of past crimes, to provide insight concerning where, and at what times, police should maintain presence, in order to make the best use of resources or to have the greatest chance of deterring or preventing future crimes.

We Indians are also greatly affected by crime directly or indirectly. As per NCRB* Data, Crimes in India have risen in 2018 as compared to last year. Criminals are becoming more sophisticated in terms of both weapons as well as technology. Hence to deal with these modern day criminals, impetus will have to given to technology in addition to advanced weapons.

Speaking of technology, this is the age of Big Data and Analytics.When data analysis and predictive modelling can be used in fields like weather ,stock markets, healthcare etc. , then why not to apprehend crime. In India, where there is a constant shortage of police force, data analytics can be a blessing .Use of advanced analytical power will immensely benefit the Police to sift through the growing volumes of data including physical records, digital feeds, social media data and others.

But, the good part is that India has also been working on this concept for quite some time now.As per a report, National Crime Records Bureau (NCRB) is attempting to use crime data analytics software to enable predictive policing so that crime can be nipped before it happens. It is expected that by post 2018, predictive policing technologies will be available to five states — Kerala, Odisha, Maharashtra, Haryana and Tripura — and then subsequently to all states and Union Territories.

Delhi Police also adopted a “predictive policing” mechanism quite some time back. The Crime Mapping, Analytics and Predictive System (CMAPS) was put in place in association with ISRO for effective use of space technology-based tools for ensuring internal security. Earlier crime mapping was a laborious process which included manual gathering and entry of records in files every 15 days. This resulted in losing a lot of vital information in the process.But the results of this combined effort are here to see.In the past one year, Delhi police has been able to save 9 different lives on 6 different occasions. There cannot be a more rewarding use of technology than to use it in saving human lives.

It doesn’t sound like a perfect fictional movie plot anymore but a reality. Data Science is influencing our lives in every possible way. Having a beforehand knowledge of when and where the criminals are going to strike , or what are their hotspots will empower our administration immensely to deal with them more effectively . However, just like everything else, it has its perils too but that in itself is a complete topic for another discussion.

A noted columnist Mark Gibbs once mentioned “This is the world of predictive analytics; the scientific version of a crystal ball. Instead of peering into a glass globe you peer into (ideally) massive amounts of data and using Big Data mining techniques such as statistics, modelling, and machine learning you look for patterns that are indicative of current or future behaviour.”

*National Crime Research Bureau

# Once Upon a Crime: Towards Crime Prediction from Demographics and Mobile Data