Structural patterns in criminal behavior.

The term crime refers to a deliberate or unintentional offense (to an individual or a community at large) that is punishable by law. although definitions in criminal law vary from one country to another (sometimes even between different jurisdictions within the same country), the taxonomy of crime (i.e., the way particular types of criminal activities are grouped together) is fairly regular and remains constant over time. Recently, many large cities around the world have launched open data portals, some publishing detailed information about every crime reported by law enforcement during the past few years, including its type, date, geographic location, and resolution. a close analysis of such data suggests that distinct criminal activities are strongly interrelated. From a network theory perspective, the different types of crimes (e.g., assault, arson, kidnapping, bribery, robbery, vandalism, narcotics, loitering, battery and theft) are viewed as nodes and the frequent co-occurrence of the same types of crime within a particular neighborhood as defining the edges of a multi-crime-type network. The aim of this proposal is to enable a better understanding of the relationships between the different types of crime by developing a mobile app that will allow local law enforcement agents to record criminal instances and classify them based on Colombia’s national penal code. Detecting structural patterns in criminal behavior at this collective level has major implication for analyzing the causes of crime, which are not identifiable by looking solely at the individual level.