The Impact of Semantic Services on Integrated Justice

Integrated justice is a life-saving miracle in theory; in practice, it’s more of a political quagmire. On the one hand are the data guardians protecting either their turf or the security of the system, perhaps both. On the other, there are the federal mandates and the needs of society. In the middle are silo after silo of disparate and sometimes conflicting data stored on far-flung systems.  In short, it’s a lot like the situation most enterprises are in.  

The difference, of course, is that human lives are at stake. Wrong or insufficient information can lead to the imprisonment of an innocent or, conversely, free a violent offender to harm again. For example, an information gap in the state of Florida contributed to the sexual abuse and murder of 9-year-old Jessica Lunsford recently, when a repeat pedophile was set free shortly after his arrest.

Mistakes have serious consequences, yet they are common in a system that requires law enforcement to check 15 or so databases, and sometimes recheck them after a finding of related data on one database, in a time frame that allows the accused to stand before a judge within 24 hours of arrest. Unfortunately, law enforcement personnel have to follow this process for thousands of individuals on any given day. Much of the data that is available is old since it takes considerable time for each agency to first receive the data and then re-enter it into yet another system. Add to that the numerous opportunities for entry errors and the willingness of criminals to provide false information and the problem compounds rather quickly.

Obviously, what needs to happen is that the right people in the justice system need up-to-date and relevant information that includes bookings, crime events and court judgments entered mere minutes ago anywhere in the country. In this way, police officers and courts can know that they are holding a violent offender, a person with an outstanding warrant from another jurisdiction, a parole violator, or a suspected terrorist and thus react accordingly. This is, of course, the goal of the Presidential mandate requiring agencies to share information.

However, it’s easier said than done. This is not a straight workflow problem where you move from point A to point B to point C, etc. Criminals simply don’t commit their crimes so conveniently and the data could be in any system with no clue for the searcher as to the parameters of time, location or modus operandi.  This is precisely the reason Semantics is an appropriate solution; because of its inference and referencing capabilities which will allow it to gather information, all the while referencing well-defined policies based on actual law, and then infer its findings to return useable, relevant information in a quick, concise and legal manner.

As to solving the data sharing problem, several efforts have already been put in place including the Joint Regional Information Exchange System (JRIES), which connects state/local law enforcement with DoD and went operational in 2003; Homeland Security Information Network (HSIN) which expanded on JRIES infrastructure and connects Homeland Security Ops Center (HSOC) with states, territories and urban areas and went operational in 2004; and, the Regional Information Sharing System (RISS), the VPN for sharing regional law enforcement information that dates back to 1974.
However, JRIES withdrew from HSIN in March 2005 due to conflicting visions and disputes over apparent Internet leaks. Today, there are multiple overlapping information sharing networks which do not quite provide all available information, nor do they provide serendipitous findings.

The strongest program out there is the FBI N-Dex program which, two months ago, introduced the Information Exchange Package Document (IEPD). The IEPD essentially defines how local and state agencies will share information with the FBI and how that information will be made available to other agencies. It is not fully delivered yet, but the process is very promising.

Interoperability issues are dealt with via several Justice Domain Standards including the Global Justice XML Data Model (GJXDM); the National Information Exchange Model (NIEM), which uses RDF as a foundation; and, the Justice Information Exchange Model (JIEM).

Data is migrating from GJXDM to NIEM, largely to enable agencies to qualify for federal funding that requires NIEM compliance. NIEM is a partnership between the Department of Justice and the Department of Homeland Security. Oddly, the Department of Defense is not part of the NIEM model. RDF is key to the vision of NIEM which encompasses other domains including transportation, intelligence, trade, and emergency management.
JIEM is a process model for the exchange of justice data between multiple agencies and is a tool put out by It has defined process models for each of the steps in the justice system, and it’s been refined to some degree, but at the end of the day, it is just a model. What we’ve done at Metatomix is apply that model to a Semantic integration platform to allow information exchanges and processes to happen. We provide frameworks for specific industries, such as Justice, incorporating all our best practices and using industry standards to allow users to build applications of their own. We also provide off-the shelf functionality, such as our Judicial Inquiry System (JIS); Judicial First Appearance (JFA, which is a real-time risk assessment program; Criminal Investigation (CI), which is an extension of JIS specifically for law enforcement; and the Judicial Data Exchange (JDX), which is modeled after the JIEM model. Each module can be initiated automatically through judicial triggering processes or manually through an intuitive web interface. Each is built on the Metatomix Semantic platform which supports judicial data exchange standards and meets rigorous security requirements.

The reason for the complex mix of functionality and features is simple. In the end, Integrated Justice users don’t care about semantic technologies-they want results. The users’ primary concern is the solution, not the technology behind it. IT staff then want to make sure the solution fits within their infrastructure and plans. Therefore, it’s important to understand that Semantic technologies are a tool, not an end in and of themselves. As usual, the complete answer requires many tools combined specifically for the task at hand.


To view the full presentation along with audio narration by the presenters, click this button. The presentation will appear in a separate window in your media player.


A case study of semantic

A case study of semantic justice in Spain: