criteo/CriteoClickLogs

Criteo 1TB点击日志数据集包含数百万条展示广告的特征值和点击反馈信息，主要用于点击率（CTR）预测算法的基准测试。数据集包含24个文件，每个文件对应一天的数据。数据集由24天的Criteo流量构成，每行代表一次由Criteo服务的展示广告，第一列表示广告是否被点击（1表示点击，0表示未点击）。数据集包括13个整数特征和26个分类特征，其中一些特征可能包含缺失值。数据按时间顺序排列，字段之间用制表符分隔。

The Criteo 1TB Click Logs Dataset contains feature values and click feedback for millions of display ads, primarily used for benchmarking algorithms for clickthrough rate (CTR) prediction. The dataset consists of 24 files, each corresponding to a day of data. It is constructed from 24 days of Criteo traffic, with each row representing a display ad served by Criteo. The first column indicates whether the ad was clicked (1) or not clicked (0). The dataset includes 13 integer features and 26 categorical features, with some features possibly containing missing values. The data is chronologically ordered and columns are tab-separated.