To keep track of intents that could be useful for a v2 / v3 of the bot, it would be useful if we could keep track of the expressions that people are asking the bot, but that we don't have a reply for yet.
Currently, the way to do this is to either keep them in the Train tab (not delete them), but then they get in the way and it's not a very clean solution. Another way is to store this separately outside of chatlayer - which is also not ideal.
It would be great if there was a "Backlog intent" option in Train, that we could assign all these expressions to, and which is not part of the NLP model. For a next bot release we can then analyse this backlog to see which intents are most wanted by the users!
Please sign in to leave a comment.