The POLE data model – Person, Object, Location, Event – is a great fit with graph databases and graph algorithms to help security and investigative teams operating in areas such as policing, anti-terrorism, border control and social services, according to Emil Eifrem.
Put simply, graph databases are designed to treat relationships as first class citizens in the data model, making it simple to join up the dots in large amounts of seemingly random data. Graph database technology has been central to a number of global investigative projects such as the Panama and Paradise Papers where it was used to mine enormous datasets at scale.
Graph database technology is a powerful enabler at effectively spotting criminal activity such as uncovering fraud rings and uncovering patterns to break up organised crime. Providing these insights based on data connections is an invaluable way of supporting law enforcement agencies, social services and other government departments in their fight against crime.
A decade ago, a G8 country’s immigration authority, for example, adopted graph database technology to allow it to visualise relationships and connections to help them work more effectively with individual cases that had been flagged up by border control officers. It found that knowingly hidden connections stood out when viewed via a system created to manage connected big data. This allowed the team to run real-time queries to spot criminal networks and fraud rings. Something that previously had been complex and extremely time consuming.
Graph database technology is also being explored as a way to enable a highly responsive informal learning system including social media, designed to support rapid decision-making.
The concept is centred around how people are connected. If one person has come to the attention of the authorities, who are they connected to and are they worth monitoring. They may be in a relationship with someone who has previously been convicted for fraud, for example. These insights can be used to support ongoing criminal investigations or start new ones based on findings.
This level of complexity is very hard to capture through conventional database technologies. Graph database technologies have been designed specifically to mine this connected data and visualise the connections.
Taking POLE position
Law enforcement agencies are examining the use of the POLE (Person, Object, Location, Event) data model for working with crime data.
Graph database technology and graph algorithms’ ability to join the dots and find connections in large amounts of data makes it a natural fit for POLE, which can be extended even further by linking in data visualisation using software tools such as Bloom.
We recently took a public dataset of a one month’s worth of street-level crime in Greater Manchester, for example, and linked it to a number of data sources ranging from geotagging data to addresses to randomly generated personal data to see how intricate a picture of these connections we could create. We soon built a database of 29,000 crimes in 15,000 locations, generating a staggering 106,000 relationships between the nodes. This test [https://www.youtube.com/watch?v=CK-QCYAFmx0] shows the sheer power that graph technology can bring to POLE investigations, maximising resources – especially where policing departments are under budgetary constraints. Together they can quite literally re-shape criminal investigations on large and small scales.
There is little doubt that graph database technology and crime data can work together to drive data driven investigations and help law enforcement authorities, police forces and government agencies to better protect the public in a world where crime and security threats are on the increase.
The author is co-founder and CEO of Neo4j, the world’s leading graph database (http://neo4j.com/)