IBM donation furthers the field of computational criminology

Husband and wife criminologists examine factors leading to urban crime

A husband-and-wife team of criminologists is studying how city characteristics such as road layout, shopping mall hours and land use changes affect the frequency and severity of urban crime in a new project aided by $5 million worth of technology donated by IBM.

Patricia and Paul Brantingham of Simon Fraser University in Vancouver, British Columbia, say the project will take the emerging field of computational criminology to a new level of complexity.

“Modeling [the factors influencing crime] requires being able to handle a lot of different data sources, being able to link them together,” Patricia Brantingham said in a phone interview. “Until IBM came along we haven’t had the computing power to be able to do this at more than a theory level and small case studies.”

The Brantinghams have been at Simon Fraser since the late 1970s. Patricia holds a university professorship in computational criminology, a term that she coined, she says. Her husband is a lawyer and professor in crime analysis.

The Royal Canadian Mounted Police is giving the couple eight years' worth of crime data from British Columbia. The Brantinghams will compare the data to road networks, transportation stopping points, new housing developments and locations of daytime and nighttime centers of activity, all the things that make up a city structure or “urban backcloth.”

What they find out could help police deploy resources and help city planners design urban settings in ways that reduce crime, says Brantingham, who has a background in mathematics and a doctorate degree in urban planning. Architectural sites can be designed to reduce what police call “crimes of opportunity,” in which property such as automobiles are made easily available to criminals.

“What we are doing is looking at the crime patterns in cities as they relate to how cities grow and change,” Brantingham says. “It makes it much easier to get engaged in proactive activity, to help reduce crime by reducing crime attractors or crime generators. It also makes it possible to pass information on to other parts of the system to work with many crime prevention techniques.”

IBM is donating $5 million worth of software, hardware and services to integrate and implement the donated equipment. This will help create a lab expected to be operational within two months and to be producing research by the summer.

Some of the software, such as IBM’s Crime Information Warehouse and Entity Analytics Solutions, is also being used by the New York City police department, says Paul McCullough, a business executive for IBM Canada’s public safety and defense sectors.

Entity Analytics was originally developed for the casino industry to uncover relationships that can be exploited fraudulently for profit, such as relationships between dealers and gamblers.

At Simon Fraser, the software may be able to find connections to gangs or relationships between criminals and locations, phone numbers or objects like stolen cars, McCullough says.

Crime Information Warehouse is a reporting tool that provides aggregate views of crime, allowing researchers to determine patterns based on locations and other factors.

Global Name Recognition is another piece of software donated by IBM that recognizes different spellings of names, particularly when there are numerous ways to spell the name of someone from a foreign country.

“What this does is take each name down to a root, and those roots can be compared to each other,” McCullough says.

The software also uncovers aliases by using algorithms that recognize the common techniques people use when they adopt a new name.

“One of the things we find is when someone obscures their name they tend to do it in a fairly predictable fashion,” he says.

IBM also donated its Information Server, a data integration platform; DB2 Data Warehouse Edition; and OmniFind Enterprise Edition, a search application.

Copyright © 2007 IDG Communications, Inc.

The 10 most powerful companies in enterprise networking 2022