The Australian Taxation Office will use a graph database from US-based startup TigerGraph within its new big data platform to aid in curbing tax avoidance and fraud.
TigerGraph has revealed its partner, Sydney-based Intech Solutions, has recently begun implementing the graph database and analytics solution at the national revenue agency.
Intech Solutions scored the $1.6 million deal in December 2020, according to the Commonwealth procurement website AusTender.
The graph database – which will be used to represent complex networks – is expected to handle terrabytes of data relating to entities (nodes) and their relationships (edges).
It will form a critical component of ANGIE, a network analytics solution designed to help the ATO’s tax avoidance taskforce discern complex, multi-layered relationships.
ANGIE – also known as the automated network and grouping identification engine – will allow the taskforce to visualise and analyse “large and complex networks of relationships”
It will do this using data from the ATO’s Teradata enterprise data warehouse and Cloudera enterprise data hub.
TigerGraph Asia Pacific and Japan vice president Joe Lee told iTnews graph databases were ideally suited to detecting fraud, money laundering and other trading irregularities.
“In graph, everything is about relationships. It’s about connections to people or connections to objects,” he said
Lee said the solution would give the ATO the ability to uncover relationships 15-20 connections deep, whereas traditional databases were “only able to connect three steps”.
“[Fraudsters] do things that are about eight connections deep to 10 to 15 to 20, so they’ll literally move the connections and the relationship about 20 levels deep,” he added.
Lee, who described the ATO’s move as “innovative”, added that Gartner had predicted the use of graph databases in analytics to climb from 10 percent in 2021 to 80 percent by 2025.
“We’re delighted to have the opportunity to support the ATO’s efforts to tackle tax avoidance and tax fraud,” Lee said.