Knowledge Grafts

knowledge grafts can be used to imbed domain knowledge in text. You can even train deep, deep learning models with some basic domain knowledge using a knowledge graph. The hardest thing about every single one of them is when the knowledge graft gets too big and how to validate it. There's just a lot that can go wrong when it gets big. Anybody else ve any thoughts or insights on eknowledge grafts? If if i may, i've not had opportunity to use them other than in code but maybe this will help gena. It seems like they could be automated on the output side or manually derived on the input side. Do you think this would be something might be able to