predictive policing

In the early 1800s to1980s, patrolling and investigating were the traditional methods that the police force were using to implement crime fighting. Uniformed police officers would patrol the streets to prevent crimes that were in progress and to catch criminals. However, researches since the 1960s has shown the limits of both patrolling and investigating for reducing crimes. This method did not show any effectiveness on crime preventions and reducing criminal activities. Police patrolling still remains the number one means of police operations but it takes up most of the resources of the police agency. On patrol, a police officer makes routine stops through designated areas. Studies have indicated that these patrols are very costly and do not reduce crimes. Because crime is not evenly distributed throughout a community, which means some places needed more patrolling than others. Facing budget cuts and not being able to hire more officers to patrol the streets, Police departments across the country are trying to find ways to combat crime. In 1994, the New York City Police Department adopted a law enforcement crime fighting strategy known as COMPSTAT (COMPuter STATistics). COMPSTAT provided a road map to getting officers back to fighting crime. COMPSTAT uses Geographic Information Systems (GIS) to map the locations of where crimes occur, identify “ hotspots”, and map problem areas. COMPSTAT has embodied an enormous amount of historical crime data. Mathematicians have designed and developed algorithms that run against the historical data to predict future crimes for police departments. Researchers based these studies on information that the police officers already know. For example, when people are victimized, the chances are that they or their neighbors will be victims again are highly increasing. Offenders usually commit crimes in the areas that they are most familiar with. This is common knowledge through the studies of criminals and criminal intents.