Phone
Phone

Crime analysis can occur at various levels, including tactical, operational, and strategic. Crime analysts study crime reports, arrests reports, and police calls for service to identify emerging patterns, series, and trends as quickly as possible. They analyze these phenomena for all relevant factors, sometimes predict or forecast future occurrences, and issue bulletins, reports, and alerts to their agencies. They then work with their police agencies to develop effective strategies and tactics to address crime and disorder. Other duties of crime analysts may include preparing statistics, data queries, or maps on demand; analyzing beat and shift configurations; preparing information for community or court presentations; answering questions from the public and the press; and providing data and information support for a police department's CompStat process. To see if a crime fits a certain known pattern or a new pattern is often tedious work of crime analysts, detectives or in small departments, police officers or deputies themselves. They must manually sift through piles of paperwork and evidence to predict, anticipate and hopefully prevent crime. The U.S. Department of Justice and the National Institute of Justice recently launched initiatives to support “predictive policing”, which is an empirical, data-driven approach. However this work to detect specific patterns of crime committed by an individual or group (crime series), remains a manual task. MIT doctoral student Tong Wang, Cambridge (Mass.) Police Department CPD Lieutenant Daniel Wagner, CPD crime analyst Rich Sevieri and Assoc. Prof. of Statistics at MIT Sloan School of Management and the co-author of Learning to Detect Patterns of Crime Cynthia Rudin have designed a machine learning method called “Series Finder” that can assist police in discovering crime series in a fraction of the time. Series Finder grows a pattern of crime, starting from a seed of two or more crimes. The Cambridge Police Department has one of the oldest crime analysis units in the world and their historical data was used to train Series Finder to detect housebreak patterns. The algorithm tries to construct a modus operandi (MO). The M.O. is a set of habits of a criminal and is a type of behavior used to characterize a pattern. The data of the burglaries include means of entry (front door, window, etc.), day of the week, characteristics of the property (apartment, house), and geographic proximity to other break-ins. Using nine known crime series of burglaries, Series Finder recovered most of the crimes within these patterns and also identified nine additional crimes.[2] Machine learning is a tremendous tool for predictive policing. If patterns are identified the police can immediately try to stop them. Without such tools it can take weeks and even years of shifting though databases to discover a pattern. Series Finder provides an important data-driven approach to a very difficult problem in predictive policing. It’s the first mathematically principled approach to the automated learning of crime series.[3] Sociodemographics, along with spatial and temporal information, are all aspects that crime analysts look at to understand what is going on in their jurisdiction.[4] Crime analysis employs data mining, crime mapping, statistics, research methods, desktop publishing, charting, presentation skills, critical thinking, and a solid understanding of criminal behavior. In this sense, a crime analyst serves as a combination of an information systems specialist, a statistician, a researcher, a criminologist, a journalist, and a planner for a local police department. This article uses material from the Wikipedia article "Crime_analysis", which is released under the Creative Commons Attribution-Share-Alike License 3.0.