Share your results -- which emotions did Resonance detect correctly or incorrectly?

[wpferrell]

Hi! Thanks for using Resonance.

We're collecting real-world results to improve the model. If you've used it, we'd love to hear from you.

Please share:

• Your use case (what kind of text are you analysing?)
• Which emotions the model detected correctly
• Which emotions it missed or got wrong
• Your dataset type (social media, customer reviews, clinical notes, other?)
• Language (English only, or multilingual?)
• Any edge cases that surprised you

Known limitation: The shame class has low F1 (0.15) due to limited training data. If you have shame-labelled data and are willing to share, please reach out.

You can also open a Discussion for questions or ideas.

Thanks!