Black-Box Emotion Detection: on the Variability and Predictive Accuracy of Automated Emotion Detection Algorithms

The current research demonstrates considerable variability in predictive accuracy across major emotion detection systems (such as Google ML or Microsoft Cognitive Services) with lower (higher) classification accuracy for negative (positive) discrete emotions. We provide two modelling strategies to improve prediction accuracy by either combining feature sets or using ensemble methods.



Citation:

Francesc Busquet and Christian Hildebrand (2020) ,"Black-Box Emotion Detection: on the Variability and Predictive Accuracy of Automated Emotion Detection Algorithms", in NA - Advances in Consumer Research Volume 48, eds. Jennifer Argo, Tina M. Lowrey, and Hope Jensen Schau, Duluth, MN : Association for Consumer Research, Pages: 831-835.

Authors

Francesc Busquet, University of St.Gallen
Christian Hildebrand, University of St. Gallen



Volume

NA - Advances in Consumer Research Volume 48 | 2020



Share Proceeding

Featured papers

See More

Featured

B9. The Power of Self-Effacing Brand Messages: Building Trust and Increasing Brand Attitudes

Tessa Garcia-Collart, Florida International University
Jessica Rixom, University of Nevada, Reno

Read More

Featured

J2. Consistence vs. Variety: The Effect of Temporal Orientation on Variety Seeking

YUAN ZHANG, Xiamen University
SHAOQING ZHANG, Quanzhou Normal University

Read More

Featured

D8. Why Employees Communicate Positive eWOM on Social Networking Sites: Motivations and Moderators

Jing Zhang, 华中科技大学管理学院
Ya Zhang, 华中科技大学管理学院

Read More

Engage with Us

Becoming an Association for Consumer Research member is simple. Membership in ACR is relatively inexpensive, but brings significant benefits to its members.