法国机动车第三方责任数据集
收藏资源简介:
本数据集名为‘法国机动车第三方责任数据集’,由拉瓦尔大学精算学院创建,包含412,748条记录,涉及一年内法国机动车第三方责任保险政策。数据集包含八个解释变量,包括暴露度、车辆功率、车龄、驾驶员年龄、车辆品牌、燃料类型、地区和密度。该数据集用于测试和比较三种不同类型的生成对抗网络(GANs)在合成保险数据方面的能力,旨在解决保险数据因隐私问题难以公开的问题,通过生成不含敏感信息但结构与原始数据相似的合成数据。
This dataset is named 'French Motor Third-Party Liability Dataset', and was created by the Faculty of Actuarial Science at Laval University. It comprises 412,748 records related to French motor third-party liability insurance policies over a one-year period. The dataset includes eight explanatory variables, namely exposure, vehicle power, vehicle age, driver age, vehicle brand, fuel type, region, and density. This dataset is used to test and compare the capabilities of three distinct types of Generative Adversarial Networks (GANs) in synthesizing insurance data. It aims to resolve the issue that insurance data is difficult to publicly release due to privacy concerns, by generating synthetic data that contains no sensitive information but retains the same structural characteristics as the original dataset.

- 1Synthesizing Property & Casualty Ratemaking Datasets using Generative Adversarial Networks拉瓦尔大学精算学院,魁北克,加拿大 · 2020年



