Kaggle Automobile Insurance Quotes and Sales Dataset

该数据集包含46,129个与汽车保险报价和销售相关的实例，并通过合成数据生成进行了丰富。数据集被划分为训练集（60%）、开发集（25%）和测试集（20%），同时排除了客户标识符变量。在训练逻辑回归模型时，使用了“Sale”变量。为了确保转换概率模型的适当输出，对历史观测价格应用了对数变换。该数据集的规模为46,129个实例，任务旨在优化价格弹性并确保公平性执行。

This dataset contains 46,129 instances related to auto insurance quotes and sales, which is augmented via synthetic data generation. The dataset is split into a training set (60%), a development set (25%), and a test set (20%), with customer identifier variables excluded. The 'Sale' variable was utilized when training the logistic regression model. To ensure proper outputs of the conversion probability model, a logarithmic transformation was applied to historically observed prices. This dataset comprises 46,129 instances, and the task aims to optimize price elasticity and ensure fairness enforcement.