MindMiner-Binary / README.md
j-hartmann's picture
Update README.md
2cb76a5
|
raw
history blame
745 Bytes
metadata
language: en
tags:
  - roberta
widget:
  - text: Alexa is part of our family. She is simply amazing!
  - text: I use my smart assistant for may things. It's incredibly useful.

This RoBERTa-based model ("MindMiner") can classify the degree of mind perception in English language text in 2 classes:

  • high mind perception 👩
  • low mind perception 🤖

The model was fine-tuned on 997 manually annotated open-ended survey responses. The hold-out accuracy is 75.5% (vs. a balanced 50% random-chance baseline).

Hartmann, J., Bergner, A., & Hildebrand, C. (2023). MindMiner: Uncovering Linguistic Markers of Mind Perception as a New Lens to Understand Consumer-Smart Object Relationships. Journal of Consumer Psychology, Forthcoming.