# NLU detects random input with wrong intent and high confidence

**URL:** <https://forum.rasa.com/t/nlu-detects-random-input-with-wrong-intent-and-high-confidence/620>\
**Category:** Rasa Open Source\
**Created:** [August 23, 2018, 7:21am UTC](https://forum.rasa.com/t/nlu-detects-random-input-with-wrong-intent-and-high-confidence/620 "2018-08-23T07:21:04Z")\
**Posts on this page:** 1\
**Showing post:** 40

<div class="post-metadata">

**Author:** ![MikesUsingRasa](https://avatars.discourse-cdn.com/v4/letter/m/71c47a/32.png) [@MikesUsingRasa](https://forum.rasa.com/u/MikesUsingRasa)\
**Post date:** [July 27, 2022, 5:50pm UTC](https://forum.rasa.com/t/nlu-detects-random-input-with-wrong-intent-and-high-confidence/620/40 "2022-07-27T17:50:32Z")

</div>

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?

---

_[View the full topic](https://forum.rasa.com/t/nlu-detects-random-input-with-wrong-intent-and-high-confidence/620)._