I am having a similar problem. A test user purposely entered the nonsense phrase: “soap on a rope.” The classifier classified it as mood_unhappy with a 99% confidence level!
The mood_unhappy intent had the word “rope” in one example, and “rope” does not appear in any other intent. “soap” does not appear in any training examples, and “on” and “a” appear in many training examples.
So, presumably it was the “rope” token that caused the classifier to classify the intent as mood_unhappy. But why 99% confidence?
I have repeated these results with other nonsense phrases where a single word appears in only one intent, and that intent is predicted with 95% plus confidence.
Any suggestions on what I should do?