[Original article: http://www.nlectc.org/pdffiles/96242-9.pdf]

Its called a "Web-based intuitive integrated interface." But in laymans terms it's called "Coplink." What it will do is help put an end to a serious problem faced by law enforcement every day . . . the inability to exchange information about criminal cases across jurisdictions.

With the help of a nearly $1 million grant from the National Institute of Justice (NIJ), the Tucson, Arizona, Police Department and the University of Arizonas Artificial Intelligence (AI) Laboratory are making Coplink a reality. When completed, Coplink will be a mechanism that not only links databases, but also searches these databases for associations and links between people, places, and things.

Were trying to provide a way for law enforcement agencies to share information, specifically case information, and to bring it all together in an integrated system where they can also do some sophisticated types of analysis on the information, says Sgt. Jenny Schroeder, project leader at the Tucson Police Department, where the system is being tested. As a trial, Schroeders department has integrated its records management system, gangs database, and video mug shot database.

The problem with having lots of isolated systems is that you have to search multiple records and databases, Schroeder says. The first hallmark of Coplink is that it takes legacy data, puts it into a warehouse, and uses a Web-based interface to make the information available to criminal justice agencies. It doesnt imply a change in anyones operational system, it simply makes data from other agencies available.

Similar initiatives that pool information are typically in the form of a distributed database system, where in all participants maintain and control their own databases. The Coplink project, however, warehouses and integrates the data at a local or regional level. These local or regional systems can then be interconnected into large distributed law enforcement intranet or a group of extranets. As technology moves ahead, new applications can be added and the interface improved incrementally. Agencies will not have to undergo redevelopment to take advantage of the rapidly changing technology. As integration with other agencies occurs, users will continue to have the same easy interface, which will keep training costs down.

With Coplink, agencies can use data from other jurisdictions and government entities that is stored on widely differing systems and computer platforms. Jurisdictions will be able to maintain data security, yet efficiently and quickly provide vital information to partner agencies. It is an aggressive approach that requires more work in the early stages but allows agencies to use more sophisticated analysis tools later on, Schroeder says.

No tool similar to Coplink has been available previously because the technology that would foster this kind of connectivity and interoperability did not exist. In addition to NIJs support, the creation of this technology was aided by scientists at the University of Arizonas Artificial Intelligence Laboratory with funding from the National Science Foundation (NSF) and the Defense Advanced Research Projects Agency (DARPA).

Hsinchun Chen, Ph.D., a University of Arizona professor of management information systems and the AI laboratory director, says an initial $7 million in NSF and DARPA funding in part paid for the laboratorys creation of a software program that helped the Central Intelligence Agency analyze data and relationships relating to Russian technologies and their use of those technologies. Chen says that when a machinenot a personextracts the meanings and relationships between data, it is termed machine learning. This ability to analyze data and relationships is the second hallmark of the Coplink project. Developed at the University of Arizona, Coplink uses concept space, a data-mining tool that points out the relationships between objects, whether those objects are people, vehicles, organizations, locations, weapons, or crimes.

In a preliminary test case, the Tucson police agreed to help a Federal agency track down a homicide suspect, even though the agency did not know the suspects name. The agency had only a confidential informants tip that the suspect had a sister living in Tucson who several years ago had been assaulted by her boyfriend. The agency did have the boyfriends name from the complaint, which the Tucson police ran though concept space. In less than a minute, the system returned the womans name and the name of her brother.

The Coplink project is expected to yield a proof-of-concept prototype for both the database interface and concept space. Everything will be very simple, Chen says. Users wont have to memorize function keys. It will work just like the Web: point and click.

After further development and testing, Schroeder says, the plan is to link the Tucson department with the Phoenix Police Department and later to extend the project to agencies throughout the State. The near-term project goal is to share public record case information with other agencies. However, current Coplink users hope to resolve issues of legality and coordination to share more investigative and intelligence information.

When fully developed, Coplink users hope to see it made available either commercially or through a nonprofit foundation created specifically to package and sell it. We want to see it made available at a reasonable cost, says Tom Coty, NIJ project manager. The public paid for its development with tax dollars, and we want to be sure it's readily available and affordable for the users.

For more information about Coplink, contact Tom Coty at the National Institute of Justice, 2025147683; Sgt. Jenny Schroeder at the Tucson Police Department, 5207914499; or Dr. Hsinchun Chen at the University of Arizona, 5206214153