Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

真实应用场景实践问题 #427

Closed
yangnianen opened this issue Jun 28, 2022 · 5 comments
Closed

真实应用场景实践问题 #427

yangnianen opened this issue Jun 28, 2022 · 5 comments

Comments

@yangnianen
Copy link

请教一下。
案例背景:P2P 中寻找欺诈黑户,黑户1 ,正常客户 0
用异构图RGCN
节点: 信用卡,用户,IP,设备
关系:用户对信用卡,用户对ID,用户对设备,用户对用户
特征:用户:金额, 其他节点没有特征

因为我是要预测客户,所以其他节点是没有标签,也没有特征,我应该用nan还是0代替其他节点的特征和标签?

@Liwb5
Copy link
Collaborator

Liwb5 commented Jun 28, 2022

事实上,这里的信用卡,IP,设备,是不是都可以作为节点特征?

@yangnianen
Copy link
Author

信用卡,IP,设备 只有和用户的关系特征,没有节点内的特征。因为官方文档上写着,只有关系特征,也要把这个加上。
image

@yangnianen
Copy link
Author

而用户有好几个节点特征,如果我不把其他节点特征设置为空特征和任意标签,拼成和总节点一样的条数的话,就会报错说越界。

@Liwb5
Copy link
Collaborator

Liwb5 commented Jun 29, 2022

你的节点特征是浮点数吗?还是离散特征?如果是离散特征的话,建议使用https://github.com/PaddlePaddle/PGL/tree/main/apps/Graph4Rec 这个代码来训练,得到每个节点的embedding,然后再写一个分类器(例如:MLP模型)来给每个节点做分类任务。

@yangnianen
Copy link
Author

节点特征有浮点型的,因为有金额这个字段。

@Yelrose Yelrose closed this as completed Jul 6, 2022
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants