天猫复购预测Baseline

天猫复购预测 Baseline，经简单处理后的数据集。

具体见“https://tianchi.aliyun.com/competition/entrance/231576/information”

商家有时会在特定日期，例如Boxing-day，黑色星期五或是双十一（11月11日）开展大型促销活动或者发放优惠券以吸引消费者，然而很多被吸引来的买家都是一次性消费者，这些促销活动可能对销售业绩的增长并没有长远帮助，因此为解决这个问题，商家需要识别出哪类消费者可以转化为重复购买者。通过对这些潜在的忠诚客户进行定位，商家可以大大降低促销成本，提高投资回报率（Return on Investment, ROI）。众所周知的是，在线投放广告时精准定位客户是件比较难的事情，尤其是针对新消费者的定位。不过，利用天猫长期积累的用户行为日志，我们或许可以解决这个问题。

Tmall Repeat Purchase Prediction Baseline, a lightly processed dataset. Details can be found at: https://tianchi.aliyun.com/competition/entrance/231576/information

Merchants sometimes launch large-scale promotional campaigns or distribute coupons on specific dates, such as Boxing Day, Black Friday, or Double 11 (November 11), to attract consumers. However, many of the attracted buyers are one-time purchasers, and these promotional activities may bring no long-term benefits to sales growth. To solve this problem, merchants need to identify which consumers can be converted into repeat purchasers. By targeting these potential loyal customers, merchants can significantly reduce promotional costs and improve Return on Investment (ROI). It is well-known that precise customer targeting in online advertising is quite challenging, especially for new consumers. However, we may address this challenge by leveraging the long-term accumulated user behavior logs from Tmall.

搜集汇总

背景与挑战

背景概述

该数据集是天猫复购预测竞赛的基准数据集，旨在通过用户行为日志预测消费者是否会重复购买，以帮助商家精准定位潜在忠诚客户并优化营销策略。数据集包含多个CSV文件，如训练集、测试集和用户信息日志，数据量较大且经过解压和切分处理，适用于大数据课程实验和机器学习模型开发。