Delivering Rare Insights with Graph Analytics Solutions

Advanced graph analytics is emerging as a powerful tool for uncovering complex relationships hidden in massive amounts of data.

In one of the many unfortunate economic side effects of the COVID-19 pandemic, credit card fraud is on the rise, according to news reports.

Just consider headlines like these:

This a problem that is particularly acute in the United States, which is the most card-fraud-prone country in the world, according to Nilson Report data. The publication says the U.S. accounts for about a third of all card fraud, which is projected to hit $32 billion globally in 2021.

Payment card companies, banks, government agencies and other organizations are working nonstop to address the growing threat of fraud. So are technology companies, which are helping the card industry use new techniques and technologies to combat fraud. And one of these weapons is advanced graph analytics, a tool used by four of the top five global banks.

Advanced graph analytics

Graph analytics is a technique that explores connections in data and reveals insights about those connections. These insights can help businesses prevent fraud, as well as enable better product recommendations, facilitate 360 degree customer views, improve supply chains and meet other data driven goals.

On the fraud front, advanced data analytics in graph databases can detect suspicious patterns of online payment activity in ways that other database systems cannot. Using deep-link analysis, graphs can analyze thousands of customer data points and the relationships between them — such as the links between people, phones and bank accounts — to deliver fraud alerts in real time. These alerts allow businesses to work proactively to stop fraudulent behavior as it is taking place.

Here’s an example from TigerGraph, a global leader in the development and deployment of graph analytics solutions. It cites the case of a U.S. multinational investment bank that is adding graph analytics to its machine learning system to find data connections between “known fraud” credit card applications and new applications. These technologies are helping the bank identify questionable patterns, expose fraud rings and shut down fraudulent cards faster — potentially resulting in millions of dollars in annual savings.

Using graph analytics in healthcare

Graph analytics doesn’t just help us protect our financial health. It can also help us protect our actual health. Graph analytics enable data scientists and business users to identify and explore complex relationships in healthcare datasets with an eye toward improving the patient wellness journey and the overall outcomes for patients, providers and payers.

In more specific terms, by combining data from the entire healthcare spectrum — including patient electronic health records, healthcare claims, provider information and historical data — graph analytics can help providers improve the quality of care while controlling costs. Along the way, graph analytics can help providers detect and prevent waste, abuse and fraud, avoid adverse reactions to drugs by linking public and internal data and measure member satisfaction.

The numbers here get pretty amazing. A native parallel graph database, such as TigerGraph, allows data scientists and business users to dig deeply through many levels of data that might encompass billions of healthcare records and millions of members and providers to explore and analyze complex relationships. And graph analytics solutions can do it all in real time.

Getting started with graph analytics

From preventing card fraud to improving healthcare outcomes, graph analytics provides powerful tools for the data-driven enterprise. For organizations on this path, TigerGraph and Dell Technologies offer valuable resources for getting started — including jointly engineered hardware-and-software solutions.

For a closer look at this partnership, see the brief “TigerGraph and Dell Technologies: Unleashing the Power of Data.”

To learn more

For a deeper dive into graph analytics and its use in real-world applications, see these resources:

About the Author: Morten Loderup

Morten is the Solution Line Manager for the Solution CoC at Dell Technologies.  In that role he develops partner ecosystems for AI, analytics, and HPC to solve customers’ use case needs in several sectors including Manufacturing, Retail, Financial Services, Higher Education and Research Institutions.  Morten has worked with Oracle and SAP business solutions since 2004 and with AI and data analytics partners since 2015 as a North America Alliance manager and Solution Line Manager.  Morten received his BA, MBA, MCIS, and MSE from BYU, University of Phoenix, and University of Texas, Austin respectively.