Cops investigating a series of robberies of fried chicken franchises in Tucson, Ariz., got a break -- a female witness recognized one of the assailants as a man she knew as "Peanut."
Tucson police used Coplink -- a specialized software package allowing users to search different criminal databases simultaneously -- and found several people associated with the nickname Peanut. When they tied the nickname to a tattoo the witness also saw, they soon located a name and address for a suspect, who quickly found himself behind bars.
In another case, a man was found face down in a parking lot after getting run over by a car, shot and having his throat slashed. He was still alive and told the responding officers, "Shorty did it," adding that Shorty had a tattoo of the name "Caesar" on his forearm. When Tucson police ran that information through Coplink, they came up with a suspect who had been released from state prison just 24 hours earlier.
Coplink also revealed a connection between the attacker and victim -- both had been arrested several times on drug-related charges -- leading police to hypothesize the incident resulted from a drug deal gone bad. With Coplink a digital mug book was generated on a laptop and shown to the victim as he lay in his hospital bed. He fingered Shorty, who was soon sporting jail garb.
Cops in Tucson say they are excited about the system's ability to search disparate databases with one query. "It's the best thing we've had in years," said Detective Tim Peterson of the Tucson Police Department.
Connecting the Dots
Before Coplink, officers had to access several separate databases, including records management databases and any number of homegrown databases -- such as gang, court citation, jail management, sex offender and probation systems. The problem was that access required specialized key strokes or key assignments.
Coplink connects all those systems and allows access with a single query.
"It puts all of this information in one easy-to-locate place that doesn't require any specialized key strokes or key assignments or special knowledge," said Peterson. "It's like any other Windows application that I'm used to doing."
As far as potential misuse, Peterson said Coplink's administrative tools provide a quick and simple way to check on who makes queries and why.
The technology migrates data from a wide range of databases and records management systems, and consolidates it into an information warehouse for law enforcement to access.
"Records management systems were spawned out of need for crime statistic reporting to the federal government," said Bob Griffin, president of Knowledge Computing Corp. (KCC), which markets Coplink. Griffin noted that the systems have evolved into more operational databases and that all of the databases are on separate platforms.
Association-detection algorithms within Coplink also allow investigators to detect people, places and things with which a suspect has associated.
"If I'm not able to locate him at any of his previous addresses, I can run his associated people and locations and come up with additional information -- where I might begin to talk to people to find out where he might be," Peterson said.
Griffin said because an integrated and consolidated warehouse has been built, "We can now apply our relationship and detection algorithms to the data, so we can determine things like physical relationships or hidden relationships."
A recent case in which a little girl was snatched demonstrated the value of Coplink's ability to detect "hidden" relationships.
The girl and two friends were playing in a park and were approached by two men in a red car. One man said his name was "Wado," and asked if the girl would help him find his lost puppy. She got in the car with the two men and they disappeared.
Her friends told police about the red car, the name Wado and that the men were white or Hispanic. Police entered those variables into the Coplink system and found a white Ford pickup had a relationship to all those variables -- it was involved in a hit-and-run accident with a red vehicle driven by the man who turned out to be Wado.
Coplink has been deployed in nearly a dozen locations throughout the country, including Boston; Phoenix, Ariz.; Pima County, Ariz.; Ann Arbor, Mich.; Redmond, Wash.; Spokane, Wash.; Montgomery County, Md.; Henderson County, N.C.; and Huntsville, Texas.
In some locations, Coplink created regional "information hubs," such as in Polk County, Iowa, which is linked, via Coplink, to seven surrounding counties. The same thing is in the works in Arizona, where Tucson and Phoenix will link to San Diego, Calif.'s Automated Reporting Justice Information System (ARJIS), according to Griffin.
"We're taking the information that's in the ARJIS database, importing it into Coplink and attaching that node to Phoenix, Tucson and Pima County, so we'll have all the border states pretty well covered," Griffin said. "Once we've got all this information built into a Coplink node, when you are searching Coplink, you are searching across all of those disparate data sources at one time."
The immediate goal of KCC is to establish regional information sharing. The long-term goal is to develop multijurisdictional, and eventually national, information sharing.
Griffin said the average county or city can procure Coplink for about a $7,500 hardware investment. Users need a server, about 1 GB of memory for the "algorithmic crunching" and high-speed disk drives.
The base price includes one data source integrated into the system. There is an added expense for each additional data source a jurisdiction adds to its node. The pricing, however, depends on the jurisdiction's size and number of sworn officers.
Griffin said most clients begin with two to four data sources, and most data sources are criminal databases, such as gang databases. Some jurisdictions have added hunting and fishing licensees, and utility customers.
"The reason they put the utility bills in there is that people may give a lot of aliases, but not on their utility bills or hunting or fishing licenses," Griffin said, adding that KCC has yet to find a data source with which Coplink is not compatible. "We tell folks that we're data source agnostic, and we truly are.
"We've developed a set of tools that allow us to read the original data source, and then through drag-and-drop mapping, we can map the fields from the original data source into the Coplink node," he said.
Peterson said he is eager to connect with other jurisdictions and agencies in the area to demonstrate the system's potential.
"It allows such flexibility in searches that it can present an investigator with investigative leads that they wouldn't have had otherwise," he said.