Crime Analytics: Predicting and preventing crime using Big Data and Analytics.

Imagine a world in which we could predict crime and police officers would be sent to the incident spot as the crime is happening, or even before it happened. Big data analytics techniques are used to solve numerous problems; Insurance companies use it to facilitate more accurate pricing, industrial plants and retail chains use it to improve production rates and increase sales. Now police and security specialists are turning to analytics to catch criminals and also prevent certain crimes from happening. This is being achieved by predicting situations based on data trends in crimes.

Patterns in criminal activities:

Just like any other human activity even crime also has patterns; while it may be difficult to predict individual behaviours, analytics can predict with considerable accuracy reactions of criminals in a specific region. Police departments all over the world are realizing these facts and are utilizing the power of analytics to assist them in their fight.

Successful implementations around the world:

Police departments of Seattle, New York and Los Angeles have incorporated big data and analytics in their efforts and all of them have reported a significant reduction in crimes, with LA recording a 33% drop in burglaries, a 21% reduction in violent crimes and a 12% decrease in property crime.

Probably the most well-known company that develops solutions for crime prediction is Palantir, who has recently stepped out of the government services arena to commercial market. Palantir’s software can work with a bunch of messy information: DNA data, audio and video recordings from different sources (including records from public surveillance cameras), register license plates to monitor car turnover, and many other global options. This solution was already used to prevent acts of terrorism in several countries. It was also used in Afghanistan to predict insurgent attacks. This software successfully located Mexican drug cartel members who murdered an American customs agent. Palantir helped to catch a child molester an hour after the abuse. Software detected him on the video recordings taken by public surveillance cameras of New York City.

Predictive policing in India:

Even the Delhi police is now turning to data and analytics to help them predict a crime and help them prevent it before it occurs. The techniques that are being used by them include data mining and using data visualization to understand and analyse trends and patterns in the occurrence of crime, helping them take proactive steps to prevent the crime from happening.

Identification of a crime hot spot and possibility of any repeat of the crime in the area through a new software called CMAPS (Crime Mapping Analytics and Predictive System). The web-based software accesses real-time data from Delhi Police’s Dial 100 helpline and, using ISRO’s satellite imageries, spatially locates the calls and visualises them as cluster maps to identify crime hot spots.

On the basis of the data, the system devises the appropriate time and place for police patrolling. Thus it results in the maximum utilization of police resources which decreases the chances of crime.