Fraud is costing banks millions of dollars. In addition to the staggering amount of dollar losses due to fraud, banks also suffer damaged reputations and the impact on their customer experience can be profound. Today, as more companies are moving their businesses online to meet changing customer expectations, they are also open to fraud attacks.
However, detecting fraud, and in particular, capturing it at the very time it is happening instead of after the fact, is no small task. Fraudsters are getting smarter by coordinating efforts and spreading their activities across large numbers of transactions in different geographical regions, making it much harder and more costly to detect unusual behavior and subtle patterns in the huge volume of transactions banks process every day.
Given the sophistication of today’s fraud rings, banks must take a different approach on how they perform fraud detection. With multiple people and large volume of transactions in play, a connected view of people, transaction, financial instrument, place and other factors in real-time will deliver deeper insights in the moment and allow quick identification of alarming events. In another word, the devil is in the relationship details.
Relational databases are not equipped to handle today’s fraud detection tasks. They are not designed for analyzing complex relationships (despite “relational” in the name) in massive amount of transactions quickly enough to prevent fraud. To query sophisticated relationships in a relational database, you will have to write a SQL query to join many disconnected tables manually, which can be extremely time consuming and requires increased computing power to return the results in timely fashion. Real-time performance is also critical for detecting fraud. For databases or other batch type data management platform unable to provide real-time performance, by the time the anomalies are identified, the damage is already done. As the volume and complexity of the transaction grow, relational database simply can’t handle the data at scale and will quickly break architecturally and economically.
This is where a real-time graph database comes in. A real-time graph database is designed for storing, managing, and querying data that has complex and/or highly connected relationships, in real time. In a graph database, both the entities (vertices) and the relationships (edges) among them are explicit, you can easily view both records and the linkages among them without joining multiple disparate datasets, making it easy to spot hidden issues and suspicious patterns.
Based on Apache TinkerPop™, the most popular open source graph computing framework, DataStax Enterprise Graph (DSE Graph) is a real-time, scalable graph capability of DataStax Enterprise (DSE), the always-on data management platform for cloud applications. With DSE Graph, banking organizations can quickly and effectively reveal the complex relationships in large volume of highly connected data the moment it’s being generated. Such deeper and timely understanding of the customer’s behaviors helps spot fraudulent transactions instantly. Unlike other graph databases, DSE Graph is part of DSE multi-model platform and is tightly integrated with DSE, which means it inherits all the enterprise capabilities in DSE, including horizontal scalability, continuous availability, and enterprise security features, all are critical for supporting fraud detection applications in the “right now” economy. DSE Graph is also seamlessly integrated with DSE Search and DSE Analytics – both are advanced features of DSE for delivering an integrated, holistic user experience.
9 out of the top 15 global banks are powered by DataStax. Attend this joint webinar hosted by DataStax and our partner Expero to learn how you can effectively combat fraud, increase your customer loyalty and deliver exceptional customer experience with real-time graph capabilities from DataStax.
Watch this on-demand webcast hosted by DataStax and our partner Expero to learn how you can effectively combat fraud, increase your customer loyalty and deliver exceptional customer experience with real-time graph capabilities from DataStax.
SHARE THIS PAGE