Statistical model predicts where a burglar will likely next strike

Anyone who slogged through a statistics class in college knows how seemingly weird, even unfathomable the science can be.  Regression analysis.  I damn near choked to death trying to get through that section.  If economics is the dismal science, then statistics is the abysmal science.  I truly admire people who get it, and who use it to create good rather than migraine headaches in undergraduates.

One new use is in identifying burglars.  Very few burglaries are ever solved, making it a rather safe field of endeavor for a career criminal.  One reason so few burglars are caught is the sheer hard work of examining stacks of crime scene reports looking for similarities between them in the hopes of identifying a pattern that leads to the culprit(s).  I doubt any police department has the staff to assign experienced, highly skilled officers to sit at a desk dredging through the reports of unsolved crimes.  About as miserable a job as a producer, or more likely an assistant, reading through the sludge pile of submitted scripts.

So along comes two statisticians who develops a model with the "easy-to-use" software to support it that does the slogging of dozens of highly skilled investigators.  This software promises to identify a series of similar crimes and identify the perp who committed them, and predicts with a reasonable certainty where a burglar will strike next, reducing a police officers job to waiting for the bad guys in a specified area.

Here's the report with a link to the original report in the attribution.
*  *  *  *  *


Statistical model helps police identify 
crime series; speeds apprehension of perpetrators

A statistical model and easy-to-use software developed by Michael D. Porter and Brian J. Reich, both professors of statistics at the University of Alabama, is designed to help local police identify a series of related crimes and nab a suspect has been unveiled.

Crime linkage is the investigative process of identifying a crime series -- a group of crimes committed by the same person or group of people. The goal of the crime linkage model and resulting software is to help crime analysts more quickly and easily sift through massive amounts of crime data to accurately discover patterns that could indicate a crime series such as one of the most difficult crimes to solve: burglaries.

Crime linkage is the investigative process of identifying a crime series--a group of crimes committed by the same person or group of people. The goal of the crime linkage model and resulting software is to help crime analysts more quickly and easily sift through massive amounts of crime data to accurately discover patterns that could indicate a crime series such as one of the most difficult crimes to solve: burglaries.

Criminology
(Classic Reprint)

by Maurice Parmelee

Order new or used from
Powell's Books
"Establishing that a set of crimes is attributable to a common offender or set of offenders is a critical first step to bringing an end to the crime spree and apprehending the criminal or criminals," said Porter, assistant professor of statistics at the University of Alabama.

Currently, police investigators must manually sift through hundreds of cases and use their expert judgment to determine whether a series of crimes is linked. Even while using query-enabled databases it is a complicated task, requiring the crime analyst to make complex comparisons of the similarity and distinctiveness of crime scene characteristics, the criminal's behaviors and any eyewitness descriptions.

The crime linkage model complements an analyst's experience and knowledge by enabling the investigator to pool information from various crime scenes, strengthening the case against a serial offender and aiding in profiling the criminal. "The model offers the potential to put linkage analysis on a more reliable and scientific basis, increases standardization, reduces the workload of police investigators and improves the prospect of using linkage analysis as evidence in legal proceedings," explained Porter.

It also can predict the likely location of the next event in an offender's crime series, an important aspect of tactical crime analysis. It does so by weighting more recent events stronger to capture changes in the criminal's behavior--namely site-selection and crime-scene behaviors.

In the course of planning and carrying out a crime and in response to the situations encountered, an offender will make a series of decisions -- perhaps unknowingly -- resulting in a unique set of behaviors, commonly known as modus operandi. Using statistical learning and data-mining algorithms, the model uses the measurable results of these decisions, recorded in a criminal incident database, to determine if crimes are linked.

Porter and Reich used 2001-2006 data on 10,670 burglaries in Baltimore County, Maryland, to build and test the efficiency of their crime-fighting model. The results show the model was highly effective in identifying crime series, including the following results:

  • The model captured 85% of links with only 5% false positives.
  • The model has a precision of 91%, indicating 91% of the 100 highest-ranked pairs are true linkages.
  • It identified 74% to 89% of true additional crimes from the series in a list of the top 50 crimes. When two or three crimes were already in the series, the model's performance improved to 86% to 91%.
  • A typical breaking-and-entering crime series was committed within 8.3 square miles and a period of 266 days.
  • Analysis of 590 crime series of four or more crimes indicates performance of next-event prediction varies greatly. 
  • If 10 square kilometers are monitored for the next breaking-and-entering crime in a series, the model can detect an average of 65% of the crimes.
Data from the Federal Bureau of Investigation estimates there were nearly 2.2 million burglaries in the country in 2010 and arrests were made in only 12.4% of the cases. The effectiveness of the model proves it is a promising crime-fighting tool that will be helpful to police investigators as they work to solve more burglaries, said Porter.

To encourage use of the model, Porter and Reich coded the model's algorithms in the R programming language and have made the software freely available to police departments, other law enforcement agencies and criminologists. "It is our hope the law enforcement and the crime research communities will use the model to combat and reduce serial crimes, such as burglaries," concluded Porter.

Related stories:
Story Source:  Materials provided by American Statistical Association.  "Statistical model helps police identify crime series; speeds apprehension of perpetrators."  ScienceDaily, 11 August 2015

Comments

Popular posts from this blog

Perfectionism a Major Factor in Suicide

The 2014 Ig Nobel Prizes: The friction of banana skins, Jesus on toast, Baby poop in sausages and more

Here, kitty, kitty, kitty. Humans met sabre-tooth cats 300,000 years ago