Why Graph Databases Are Ideal for Fraud and Compliance

The advent of graph databases is powering a new generation of applications that require modeling interconnectedness in a more natural and easier manner, resulting in exponential improvements in performance and adaptability. Graph databases are ideal for traceability applications built on connected relationships between hundreds or millions of different entity types. The most well-known interconnected application is LinkedIn, where individuals are connected to each other and form their own personal network -- leading to the adage that every actor in Hollywood is connected to Kevin Bacon within six degrees.

Coupled with entity resolution technology, graph databases can power the new generation of interconnected traceability applications such as ERP, supply chain management, fraud detection, compliance, target marketing, recommendation engines, master data management, digital asset management, and network management for Telecom, IT, power grids, and infrastructure.

The Limits of Relational Databases

Previous traceability platforms using traditional relational databases have proven to be limited, particularly as systems need to expand to track large numbers of disparate entities. As the number of inter-relationships between data tables grows, so does the complexity of queries. With both the number of records and the number of relationships exploding, system responsiveness is severely impacted by multiple JOINs between tables. Relational databases also have fixed schemas, so they do not adapt well to changes. A small change to one table can cause a ripple of changes across the entire system that must be carefully accounted for. A seemingly simple change like adding or replacing a column in a table can become a million-dollar task in a large relational system.

Why Graph Databases Win

Graph databases are designed specifically for recording, defining, searching, and visualizing genealogy linkages between assets, owners, places, and events. They offer numerous advantages over relational databases in traceability applications:

Performance at Scale

Graph databases support millions or hundreds of millions of records without impact to performance, as the data traverses across the various entity relationships. The query speed does not degrade as the number of relationships grows -- which is the opposite of what happens with relational JOIN operations.

Flexibility

New inspection points, new value-chain participants, or new compliance points can be added across the workflow without significantly impacting the database. This is the core structural advantage over relational databases, where schema changes are expensive and risky.

Security

Graph databases can provide the highest level of user and data security, allowing data contributors to see only their own data contributions and no other data -- delivering security and privacy suited to multi-party compliance and supply chain applications.

Natural Visualization

Graph data can be interpreted intuitively using node-link visualization, representing the full path between source and destination nodes and all the interconnected entities between them. For fraud and compliance applications, this means an investigator can visually trace the exact path of a transaction or identity through a network of connected entities -- something that would require complex SQL queries and exported data to accomplish in a relational system.