Abstract
Due to its emerging nature, cyber risk is a field in which very few data exist when it comes to calibrating models for the severity of such events. This makes the task of actuaries particularly challenging regarding pricing or claim management. In the present note, we present a general methodology to process text data using neural networks, and how it can be used to determine the severity of a cyber incident when this information is missing. This methodology is illustrated on a public benchmark database. It can be used either to augment database, by adding some claim incidents to historical databases, from reported incident whose cost is unknown. It can also be used to quickly evaluate the severity of a cyber claim just after its occurrence.
Keywords: Cyber risk ; Natural Language Processing ; Neural Networks.
Sector: Insurance
Expertise: Cyber risk
Authors: Justin Kher,
Olivier Lopez and Hugo Rapior
Publisher: Detralytics
Date: May 2023
Language: English
Pages: 33
Reference : Detra Note 2023-4
About the authors
Justin Kher
Justin holds a Master’s degree in Mathematics (Actuarial Route & DU in Actuarial Science) from ISFA in Lyon, France. In the past years, he has gained experience in inventory, provisioning and automation. He has also worked on several projects such as the estimation of the severity of property and casualty claims by Random Forest.
Olivier Lopez
At Detralytics, Olivier is a Scientific Advisor and he coaches young talents, provides cutting-edge training, fosters innovation and oversees R&D projects.
Olivier is also a professor at ENSAE in Paris (FR) and also co-director of the Master 2 in Statistics, Finance and Actuarial Sciences. He is a Qualified Actuary and has a compelling academic background developed in prominent universities.
Hugo Rapior
Hugo is part of the Talent Consolidation Program (TCP) and is also People Lead at Detralytics. Before joining Detralytics, Hugo worked in life and non-life insurance for a major player in the insurance industry, within an international division. Since joining the firm, Hugo has had the opportunity to extend his skills in non-life pricing, across a range of product types (Motor, MRH, Bank Cards and Cyber).
