search

dingjingtao / SRNS

Our implementation of neurips'20 paper "Simplify and Robustify Negative Sampling for Implicit Collaborative Filtering"
34 stars 7 forks source link

preprocess of dataset #2

Open JianhuanZhuo opened 3 years ago

JianhuanZhuo commented 3 years ago

Your work is awesome. I wonder if there is any preprocess of datasets. Why the ML-1m only contains 563,186+6,028+6,028 instance rather than 1m originally.

QuanYuhan commented 3 years ago

Thanks for your attention! The preprocessing method of the dataset is in Appendix C.1 of the paper. We only keep instances with higher scores and filter out users with less than 5 instances.

JianhuanZhuo commented 3 years ago

Thx you! But if it is possible to provide the E-commerce dataset? Or just tell me how and where can I obtain it? In your paper and its supplemental, I can not find any description about how to get it.