The first Scenario of the Insurance Demonstrator is aimed to exploit at best the first release of the AEGIS platform. Thanks to the cooperation between HDI and the technical partners the execution of the Scenario can be considered as a success: the current results satisfy the needs of HDI, while opening some interesting challenging point to be addressed with the next releases of the platform and further improvement of the HDI-related tools.
A key point of the first Scenario is the detection of an event from the Event Detection Tool: four types of event have been identified by the HDI Data Scientists as crucial for an insurance company, namely floods, hailstorms, whirlwinds and social events (like strikes or protests). AEGIS allows HDI to have a fast overview of the impact of the notified event on its customers’ portfolio, evaluating the number of customers that have an asset(s) insured in the area and, depending on the type of policy(-ies) owned, estimating the financial impact of the event (risk exposure) for the company. The worth of the in-house datasets is enriched with external (open and purchased) datasets related to the reliability of the HDI customers. The cross-analysis of these datasets lead to the opportunity to assign a priority value for the contacting of the customers that could have been involved in the event.
A detailed schema of the actions performed and the tools utilised for the execution of the scenario is reported below.
The main issue related to the Early Demonstrator and, in general, to the Insurance Demonstrator is the privacy and security regulations. Since the topic is crucial and it needs a deep knowledge and accurate handling, one of the Ethical Advisory Boards has supported HDI for the scenarios’ definition. In order to respect the Italian and European Legislation about data treatment, and the Insurance-specific policies, the in-house datasets stored in HDI databases have been uploaded on the platform after their anonymization through a tool provided by a partner of the AEGIS Consortium.
The Early Demonstrator has satisfied all the HDI actors, in terms of usability of the tools and accuracy of the analysis. Toward this end, the cooperation between the HDI employees and the technical team of the project was fundamental.
The execution of the scenario demonstrates the worth of AEGIS for the insurance sector: the customer-driven HDI business model is well supported by the AEGIS tools, the customers’ details are quickly displayed and allow the Data Scientists to identify those who can be involved in an event. The data enrichment provides further information, enabling the analysis of different databases to establish the rules for an efficient prioritisation algorithm.
Contacting customers near the happening of an event, on one side strengthens the relation between HDI and its customers, while on the other decreases the risk of fraud: it is not unusual that after many months, even years, from a whirlwind for instance, a customer claims for damages. In that case it is difficult for the insurance company to prove the damage was not due to that whirlwind. With a recorded call near the happening of an event, prowlers are catch unaware decreasing the fraud risk.
For further information, you may have a look at the complete description of the First Insurance Scenario Execution in the AEGIS Deliverable D5.3 – Demonstrators Evaluation and Feedback – v1, available at (https://www.aegis-bigdata.eu/public-results/).
Blog post authors: HDI – GFT