Is there a way to do both link prediction and regression task together?

Questions
#1

I would like to understand if link prediction and regression can be done together in the same model, and if so , kindly refer me to some good resources. Thank you for your time.

#2

What kind of regression problem are you considering? In theory you should be able to use a single GNN encoder for multiple tasks.

1 Like
#3

My use case would mostly require decision tree/random forest model. Maybe for deployment , I would attempt at using a regularised LR model too instead.

1 Like
#4

It sounds good to me. Feel free to update this thread if you have further questions.

1 Like
#5

Could you send along any papers with code or resources that has previously done this, if your time allows it? I am having trouble putting these tasks together. I appreciate it.

Edit: regression for knowledge graph usecase

#6

I don’t know papers that have done this before.

#7

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.