Network Chart Template for TIBCO Spotfire® 1.0

This template aims at providing you with a means to create a network chart to display the multi-directional relationships between the nodes in your data. This type of work is especially important in the context of criminal investigations: starting from a few users with suspected fraudulent behaviour, identify accomplices in their same fraud ring. Filter straight to the fraud ring, separating victims from criminals, in a small chart that milks just the right information from underlying potentially huge data sources. See some of the industries where this matters.

• anti-money-laundering (banking): identify relationships between bank accounts that are in the paths of relationships of more than one potential criminal.
• claim investigation (insurance): identify fake health providers whose receipts appear in a variety of customers.
• telco fraud (telco): identify fraud accomplices as numbers who, despite depicting innocent behaviour, are within the contact networks of two or more fraudulent detected numbers.

The intended logic is that you take a set of suspicious IDs and collect from your Big Data source all IDs that contacted with them up to k degrees of separation. We enrich the data colletion process so as to identify nodes that are in the connection path of two or more alarms. This enrichment during the data collection process before the actual production of the network chart has the potential to allow us to find connected paths faster than standard graph algorithms, as these search blindly a posteriori through the whole network. For more info on how to adapt your search to a large data problem, ask us.

This type of visualisation is also useful in other usecases. For example, within marketing, to depict the relationship between people that interact on social media, which is useful for identifying community trend-setters against trend-followers and for example aiming promotions at the first at the launch of a new product. Or depicting the relationship between word usage in unstructured data sources.