You’ve heard of machine learning being used against fraud, but how does the technology actually work? What does a model need to look at, and how is the structure organized so it can make sense of so much variable data?
Machine learning (ML) can often be seen as a ‘black box’ or used as a buzzword without delving into the reality beneath. In this guide, we go under the hood to explain how we use ML at Ravelin, to make it as simple and clear as possible.
- How we select and engineer features from data to feed into ML models
- Categorization of features - including text, network topography and more
- Reasoning behind the three key ML architectures we use: TabNet model, natural language processing model and anomaly detection model
Download your copy of the guide for free