Graph Machine Learning:

It is expected that by 2023, 30% of the organizations worldwide will already be using graph technologies to facilitate rapid contextualization for decision making, and that application of graph processing and graph databases will grow at 100% annually according to Gartner. Graphs offer a novel source of information since they accurately and adequately capture the interactions of different entities of interest such as organizations, people, devices, and transactions. In fact, it has been discovered that connections in data are as valuable as the data itself, as these provide context allowing algorithms to learn not only from the datapoint itself but also from the structure created and the flow of information. The interest in these technologies and algorithms have made the field of graph machine learning the fastest growing field in the major AI conferences. In this talk, we will briefly describe the concepts needed to understand Graph Machine Learning, describe the evolution it has taken in its methodologies, give a brief overview of the field, and finally, show real use cases of graph machine learning algorithms to detect fraudulent activities, identify potential influencers, and to enhance credit risk scores.