assume i asked the bot i want two person akela and amanda and can i extract this two names at the same time?
If I understood your question, I think you can if you provide more examples in nlu.md files to related intents.
What if am using spacy for extracting names??
I think it could help : What is the best method for entity extraction for names?
Yes I did see that before…but that question is only for one name and what I want is …is it possible to extract two names at the same time…for example.*I want to get in touch with adele and bella * can these two name be extracted within one intent or it will.just took one if am doing *name =tracker.get_slot("name ").and there is two names and how will this tracker get the name ??
Yes, you can extract multiple entities in the same sentence. The last entity of a specific type that is extracted will be used to fill a slot of the same name. You can gather all the extracted entities in a custom action by looking at the message parsing info. e.g.
all_entities = tracker.latest_message.get("entities", ) name_entities = [e for e in all_entities if e.get("entity") == "name"]
Okay I will try that…here for extracting name which way is better like duckling used to extract time,number and such things from a text or user utterance or message??
I think this is the same question you were asking here Do I have to put from_entity inorder for duckling to extract while we are doing formaction in rasa 2.0? so I’ll just quote the answer from there for the record:
Duckling does not extract names. It extracts regularly patterned entities (like dates, numbers, etc.). For proper names, try SpaCy.
No no that is not my question here…I say which way is better to extract name like duckling extract time and others…what I mean is just like duckling for extracting such entities what other extraction method is there for name…am not saying how to use duckling for name .am saying other way of extracting name that is what I meant here
And I tried spacy but still I have to put all possible names for spacy to extract
Please see my reply on the other post linked to above.
yes i did i replayed u am waiting for ur answer