关于构建自己的kg数据集

[zzzzljaa]

非常感谢贵团队开源的代码！阅读之后收获颇丰！

我们希望能在其它更多的数据集上进行实验，但是苦于没有数据集文件生成的相关说明，如datasets/kg下各个数据集生成的详细过程，因此我无法生成其它数据集的相关文件。期待贵团队回复❤十分感谢！

[Re-bin]

嗨 👋！

感谢对SSLRec的关注！我们正在准备数据集文件说明，以下是关于KG部分的，供你参考 🤗

• train.txt This file contains the observed user-item interactions for training. Each line starts with a User ID, followed by the IDs of items that the user has interacted with.
• test.txt This file contains the user-item interactions for testing. Each line starts with a User ID, followed by the IDs of items that the user has interacted with.
• kg_final.txt This file contains the KG tripltes. Each line of the record follows the format [item_id, relation_id, entry_id].
• item_list.txt This file contains the item ID information, which includes org_id, remap_id and freebase_id.
• entity_list.txt This file contains the entity ID information, which includes freebase_id and remap_id.
• user_list.txt This file contains the user ID information, which includes org_id and remap_id
• relation_list.txt This file contains the relation ID information, which includes org_id and remap_id

希望可以帮到你 :)

Best regards,
Xubin

[zzzzljaa]

感谢回复❀