Tmall dataset

该数据集来源于天猫平台，记录了用户与商品之间的互动信息。该数据集被划分为三个子集，分别用于预训练、微调和测试。其规模包括约800万个不同的行为标识和大约390万的样本。该数据集的任务是预测用户在未来5天、10天、15天和30天内最感兴趣的物品类别。

This dataset is sourced from the Tmall platform, recording interaction information between users and commodities. It is divided into three subsets for pre-training, fine-tuning, and testing respectively. It contains approximately 8 million unique behavior identifiers and roughly 3.9 million samples. The task of this dataset is to predict the item categories that users will be most interested in within the next 5, 10, 15, and 30 days.

搜集汇总

背景与挑战

背景概述

Rec-Tmall是天猫推荐竞赛提供的数据集，包含商品属性、用户行为日志和用户评论三部分数据，总数据量超过14亿条记录，适用于推荐系统相关研究。数据集规模庞大，覆盖了商品信息、用户交互行为和评价反馈等多维度信息。