Top 5 use cases for graph databases

The graph database stands as one of the greatest innovations that came out of the NoSQL database boom that shook the industry over a decade ago. Graph databases are designed to derive insights from massive amounts of interconnected data. They store relationships between data objects within the objects themselves, enabling lightning-fast analysis that is nearly impossible to achieve by other means.

Graph databases are intended to run alongside relational databases – which are still the workhorse repositories in most enterprises – rather than replacing them. Their main advantage is the ability to quickly run complex queries on data from multiple systems without the overhead of table joins or data transformations. Aggregating that widespread data requires data integration efforts, often in the form of a data lake.

The benefits of graph databases go beyond query speed. Complex relational models no longer need to be worked out in the usual tedious way because relationships can be easily modeled and schemas can change dynamically. But those who are fluent in SQL need not feel left out; graph database query languages ​​such as gsql are SQL-adjacent languages ​​supplemented with graphical capabilities.

Significantly, the emphasis on relationships and the ability to efficiently process large amounts of data make graph databases ideal for artificial intelligence and machine learning (ML) AI applications. That combination can be enhanced when the graph database software includes AI/ML specific tools and interoperability features.

So what are the emerging use cases of these new capabilities? Here’s how five industries are taking advantage of the extremely fast relational query performance of graph databases in distributed data stores.

1. Get a 360-degree view of customers

Interactions between companies and their customers or sales prospects are often complex, with many touch points. Ideally, these should yield sales strategies that continually adapt to customer needs. Such 360-degree scenarios quickly lead to many-to-many relationships that, using a relational database, would require cumbersome modeling and table linking to yield actionable insights.

This is the kind of situation where a graph database excels. For example, UnitedHealth Group (UHG) has implemented a graph database to help improve the quality of care for more than 26 million members while reducing costs. UHG is the largest healthcare company in the world by revenue, using a massive graph database to track more than 120 billion relationships between members, healthcare providers, claims, visits, prescriptions, procedures and more.

In addition to its graphical database, UHG has developed several GUI applications that, among other benefits, provide a consolidated view of member interactions between physicians, pharmacies, clinical labs, health consultants, and UHG itself. More than 23,000 users access the database daily, enabling healthcare providers to make better care and wellness recommendations in real time based on the latest member activity. UHG predicts that the cost savings could eventually run into the billions.

2. Transforming financial services with AI

The exponential growth of data has been the biggest driver of AI/ML, which requires large amounts of data to uncover meaningful patterns and improve decision-making accuracy. Few industries are more data-intensive than financial services, but as with other industries, data comes from many different sources and typically ends up in relational database silos.

By bridging these silos, graph databases can help AI/ML deliver superior predictive analytics, risk management, fraud detection, anti-money laundering, insider trading monitoring, automated customer recommendations, and more. Also, a graph database combined with AI/ML can ensure that data is clean in the first place, reconciling anomalous differences in customer records and financial product attributes that can lead to inaccurate results.

Intuit uses graph database software in conjunction with AI/ML to transform from a product company to an AI-driven expert platform company. An important part of this journey is creating knowledge charts, which enrich data and surface insights from clusters of related elements. Intuit combines knowledge graphs with the most advanced form of ML, deep learn, to power Intuit’s chatbots and in-app recommendations. Normally it is difficult to determine how deep learning leads to results; an important benefit of Intuit’s knowledge graphs is that they add “explainability” to deep learning.

3. Optimize supply chains

One of the lasting effects of the coronavirus pandemic is the realization that global supply chains can be appallingly fragile. With or without disruption, manufacturers are well aware of how complicated many supply chains are to maintain and optimize.

Think about the day-to-day challenges that automakers face. The first requirement is to accurately forecast customer demand to determine the number and types of parts to order – right down to the various models and options buyers are expected to choose. Those forecasts need to be synchronized with parts availability from hundreds of suppliers, along with manufacturing efficiency and supplier risk estimates.

Jaguar Land Rover (JLR) chose a graph database solution because it could encompass the many silos of data that needed to be tapped for supply chain analysis – and explore the matrices of relationships between data elements. The primary goals were to increase average profit per unit sold and reduce obsolete inventory, along with minimizing the effects of supplier disruptions. Some key supply chain planning questions at JLR now take 45 minutes instead of weeks and, more importantly, management can answer questions it could never have asked before.

4. Improving online retail business

Retail e-commerce companies are under increasing competitive pressure to deliver better customer experiences based on accurate customer data and purchase histories. That foundation enables everything from dynamic pricing to product recommendations to personalized special offers, all of which are based on data collected throughout the customer journey.

Graph databases can help in a number of ways. Consider the possible relationships – between customers and payment methods, customers and brands, products and return rates, promotions and sell-through rates, and much more. Let’s say you want to conduct a search to determine which promotions are most effective for a particular product when offered to a subset of customers defined as loyal. That would take a long time with a relational database, but a graph database can return the results with very little latency.

The seemingly simple act of reliably identifying which customers bought what can be enhanced by a graphical database, which can aggregate and reconcile all associated customer data, regardless of payment method or point of sale. In a three-month test of a graph database, a large e-commerce company discovered 12 million new account connections across its five different retail websites. The company estimated efficiency savings of nearly $3 million and projected a 17.6% increase in sales.

5. Improving fraud detection accuracy

We have all witnessed the evolution of fraud detection through our banking, credit card and telecom companies. Early rules-based attempts tended to overlook questionable transactions or flag innocent transactions as fraudulent. However, as the financial industry adopted graph databases to augment their AI/ML efforts, fraud detection accuracy improved noticeably.

Graph databases combined with AI/ML improve fraud detection accuracy, reduce false positives, and detect anomalies that might otherwise be overlooked. Machine learning needs to use many different data types to model a customer’s normal behavior: location, device, payment type, authentication method, and so on. In addition, what are defined as normal patterns of behavior must be immediately modified in response to legitimate change. Graph databases support that dynamic and allow AI/ML to traverse customer interactions to identify significant differences.

Financial services firms JP Morgan Chase and Intuit have both adopted graph databases to enhance their AI/ML fraud detection efforts. JP Morgan Chase uses a graph database to help protect more than 60 million homes in the US. According to Intuit, graph-based machine learning has enabled the company to detect 50% more potential fraud events and reduced false positives by about the same percentage.

These are just some of the most common uses for graph databases. Customers also use graph databases to optimize business processes, improve healthcare outcomes, sharpen digital marketing campaigns, identify cybersecurity threats and even manage energy networks. New applications are added regularly.

The mission of the graph database is to open a whole new window on relationships between data elements, delivering analytics that can identify new business opportunities, spot wasted motion, and provide an agile foundation for AI/ML initiatives. When given access to multiple corporate data stores, graph databases can provide entirely new insights and possibilities.

Yu Xu is CEO of TigerGraph.