Using Artificial Intelligence to Price Insurance Risks
Insurers rely on comprehensive historical claims datasets to price risk accurately, but claims data does not exist for new and emerging risks such as cyber attacks. As a result, large segments of human and economic activity are underinsured. Cytora is solving this problem by leveraging machine learning against web data to generate synthetic claims history, enabling insurers to price risk where they previously had zero data.
In this talk, we will show that the collection and processing of semi-structured and unstructured web-based data using AI algorithms, including neural network based sentence multi-class classification and unsupervised topic clustering, can be applied to create synthetic loss event datasets from openly available data.
We will also discuss how this approach can generate loss frequency models suitable for pricing risks like automotive recalls, factory fires and food safety incidents.
Aeneas draws on 10+ years of technical experience developing bespoke cloud-driven web applications. He holds masters degrees in Theoretical Physics and Computational Science from Imperial College London. His widely published postgraduate work focused on optical materials used for chemical sensing and invisibility cloaking. In 2011 Aeneas co-founded the Hermes Academy, an innovative nonprofit organisation backed by the Royal Society of Chemistry.