Allstate Purchase Prediction Challenge 根据交易历史预测购买的政策

当客户购买保险单时，他/她将在购买计划之前收到许多具有不同承保范围的报价。在此数据集中，这表示为一系列行，其中包括客户ID，有关客户的信息，有关报价的政策的信息以及成本。您可使用总交互历史的有限子集来预测购买的承保范围选项。如果可以在购物窗口中更早地预测到最终的购买，则可以缩短报价过程，并且发行人不太可能失去客户的业务。

When a customer is shopping for an insurance policy, they will receive multiple quotes with varying coverage options before finalizing their purchase plan. This dataset is structured as a collection of rows containing customer ID, customer-related information, information about the quoted insurance policies, and cost details. One can utilize a limited subset of the total interaction history to predict the coverage option that the customer will ultimately purchase. If the final purchase can be predicted earlier within the shopping window, the quoting process can be streamlined, and the insurance issuer will be less likely to lose the customer’s business.

搜集汇总

背景与挑战

背景概述

该数据集旨在基于客户的交易历史预测其最终购买的保险政策选项，包含训练集和测试集，其中训练集提供完整的报价历史及购买记录。数据集涵盖了客户特征、报价信息、时间地点等多维度变量，用于支持预测模型的构建。