Improving state-of-the-art emotion detection with human signals

This project brings together two topics from our previous work: On one hand, we have a built a state-of-the-art model for emotion detection in context [1]. On the other hand, we have a dataset of recorded eye-tracking and EEG signals for subjects reading text [2]. Preliminary experiments [3] and previous research [4] show that these human signals are valuable for emotion detection.

The goal of this thesis or semester project is to improve the performance of the emotion detection system by developing a method of extracting or predicting brain activity signals relevant for this task.

Supervised by Nora Hollenstein


  1. [1] SemEval 2019 - EmoContext
  2. [2] Hollenstein N., Rotsztejn J., Troendle M., Pedroni A., Zhang C. and Langer N. “ZuCo, a simultaneous EEG and eye-tracking resource for natural sentence reading.” Scientific Data. 2018.
  3. [3] Rotsztejn, J. (2018). Learning from Cognitive Features to Support Natural Language Processing Tasks (Master thesis, ETH Zurich).
  4. [4] Li, M., & Lu, B. L. (2009, September). Emotion classification based on gamma-band EEG. In Engineering in medicine and biology society, 2009. EMBC 2009. Annual international conference of the IEEE (pp. 1223-1226). IEEE.