Retailrocket

该数据集包含了用户的购买和点击行为信息。为了处理数据，每个序列的最大长度被设定为10；如果某个序列超过了这个长度，它将被分割成多个子序列；而对于那些较短的序列，则通过填充零来进行长度补充。该数据集的任务是进行序列化的推荐。

This dataset comprises user purchase and click behavior data. For data processing, the maximum length of each sequence is set to 10. If a sequence exceeds this length, it will be split into multiple subsequences; for shorter sequences, zero-padding is applied to achieve the target length. The task of this dataset is sequential recommendation.