An AI Factory team releases internal software to analyze relations

Santiago Basaldúa

d&a blog

A 283 year old field of mathematics founded by Leonhard Euler with his “Seven Bridges of Königsberg” problem is changing the way we tackle customer analytics at BBVA. Rather than studying clients, corporations, managers, products or ATMs as static objects with standard attributes or “labels” such as those emerging from traditional customer segmentation, this field, called Graph Theory, focuses on their relations. A graph, in the mathematical sense, is a set of elements and their relations.

Relations are everywhere. We create relations all the time by deciding what we purchase or where we live. Also, many problems are inherently relational, fraud in all forms, but also product recommendation or language itself can be studied using graphs.

By focusing more on data flows and behaviour and less on specific client “labels”, we very much avoid the risk of profiling and unfairness. We also find unthought-of opportunities that we can offer as meaningful recommendations generating value for our clients.

And not just financial relations, the relations of the words in a sentence can also be studied using graphs. This helps better understand word choices and produces a significant improvement in customer advice.

The Graph Analytics Team at the AI Factory has created a software library that simplifies the use of graphs to all analysts at BBVA and also includes some original algorithms created by them. Besides providing tools, the team delivers ready-to-use graph assets and takes part in internal training programs to help BBVA, as a financial technology leader, complete its path towards full maturity graph adoption.