search

haseebs / OWE

Pytorch code for An Open-World Extension to Knowledge Graph Completion Models (AAAI 2019)
https://aaai.org/ojs/index.php/AAAI/article/view/4162
37 stars 9 forks source link

Question about sampling strategy for FB15k-237-OWE #16

Closed lujiaying closed 2 years ago

lujiaying commented 3 years ago

Hi Haseeb,

I have a question regarding your sampling strategy to produce FB15k-237-OWE.

Each picked head x is removed from the training graph by moving all triples of the form (x, ?, t) to the test set and dropping all triples of the form (?, ?, x) if t still remains in the training set after these operations.

For dropping all triples of the form (?, ?, x) if t still remains. What does t refer and it seems (?, ?, x) not contains t?

haseebs commented 2 years ago

I just noticed your issue.

This seems to be a typo. It should be x instead of t. -> "...dropping all triples of the form (?, ?, x) if x still remains."

Basically, we don't want the head entity x to exist in the knowledge graph when doing tail prediction, since the head entity x is supposed to be unknown (which is why it is called an open-world entity).

lujiaying commented 2 years ago

Gotcha, thanks!