Hi, here I got a tricky issue on entity extraction, let’s say we have a statement:
“I’m short tempered and frustrated with my partner who is suffering from lack of motivation and started to drink again”
In this sentence there are two people being mentioned: “I” with entities “short tempered”, “my partner” with entities “lack of motivation”, “started to drink again”
So currently I can only train a custom CRF to pick up these entities but can’t tell which entity is related to “me" and which entity is related to “my partner”, I did some research and I reckon spacy dependency parsing can be helpful in this situation, but I am not sure if it’s the best way, just want to see if you guys have any idea on how to solve this problem?
Thanks, Patrick