+1
Under review
Ivan Gavran 2 months ago in Newbie • updated by Praveen Srinivasaiah (Admin) 2 months ago 5

Hi,


I am a new user of onsequel and the framework seems pretty good, congrats :)


My question is: when using nlp player decision node and my bot doesn't recognize the answer, as a "no match route" I'd like to add: "didn't get you, can you rephrase your answer". However, if I try to use go-to node, it only allows me to move control to some of the messages, and not to node waiting for user input. So, looking for something like nlp player decision node self loop.

What is the best way (or is there a way) to get this functionality?

Under review

You can use the NLP fallback button node type. Let me know if this helps.


Thanks for your answer. It is not what I am looking for, though - I would like to have an option to loop. So, upon going to fallback button, that I can return to entering user's input again, in the same state.

Example:

Bot: "What would you like to have?"

<state A>

User: "Lemonade"

Bot: "We don't have it. Something else?"

<state A>

User: "Green tea"

Bot: "We don't have it. SOmething else?"

<state A>

User:

...

So that the same loop keeps repeating until User asks for something that is offered.

+1

How about using a GoTo node pointing at the end of NLP node and returning back to the input node?

That is the problem - I was not able to connect end of NLP node (here: end of fallback) and input node (it would always take me one node up).

The solution I found - although it is a bit of a hack - was to add additional message node (with empty message) in between question ("what would you like to have") and the NLP choice node. Then I would use GoTo and point it towards this empty message node.

Is there a better way? Is there a way for me to share the example I was describing so that it's clearer what I am talking about?

+1

Actually, we use this method you've mentioned, like an hack with those empty nodes.