Cultural differences in emotion recognition

Publications

Quiros-Ramirez, M. A., & Onisawa, T. (2015). Considering cross-cultural context in the automatic recognition of emotions. International Journal of Machine Learning and Cybernetics, 6(1), 119-127.
Quirós-Ramírez, M. A., & Onisawa, T. (2014, October). Cultural dimension in emotion recognition for human machine interaction. In 2014 IEEE International Conference on Systems, Man, and Cybernetics (SMC) (pp. 711-716). IEEE.
Quiros-Ramirez, M. A., Polikovsky, S., Kameda, Y., & Onisawa, T. (2014, June). A spontaneous cross-cultural emotion database: Latin-America vs. Japan. In KEER2014. Proceedings of the 5th Kanesi Engineering and Emotion Research; International Conference; Linköping; Sweden; June 11-13 (No. 100, pp. 1127-1134). Linköping University Electronic Press
Quiros-Ramirez, M. A., Polikovsky, S., Kameda, Y., & Onisawa, T. (2012, November). Towards developing robust multimodal databases for emotion analysis. In The 6th International Conference on Soft Computing and Intelligent Systems, and The 13th International Symposium on Advanced Intelligence Systems (pp. 589-594). IEEE.
Quiros-Ramirez, M. A., & Onisawa, T. (2012, October). Assessing emotions in a cross-cultural context. In 2012 IEEE International Conference on Systems, Man, and Cybernetics (SMC) (pp. 2967-2972). IEEE.
Polikovsky, S., Quiros-Ramirez, M. A., Kameda, Y., Ohta, Y., & Burgoon, J. (2012, August). Benchmark Driven Framework for Development of Emotion Sensing Support Systems. In 2012 European Intelligence and Security Informatics Conference (pp. 353-355). IEEE.
Polikovsky, S., Quiros-Ramirez, M. A., Onisawa, T., Kameda, Y., & Ohta, Y. (2012, November). A non-invasive multi-sensor capturing system for human physiological and behavioral responses analysis. In IAPR Workshop on Multimodal Pattern Recognition of Social Signals in Human-Computer Interaction (pp. 63-70). Springer, Berlin, Heidelberg.

Description

Shop interactions regularly involve a shop attendant and a client, where the shop attendant tries to 'read' the client's wishes, likes and dislikes to offer further goods recommendations. In this project, we automatize the inference of the user's moods or emotions using ML and embed it in a recommender system. The key point of the project is cultural awareness. We investigated the effect of cultural background in the expression and recognition of emotions and found out that culturally dependent emotion recognition models increase the performance and user satisfaction, while culturally generic models hinder performance of the systems in which they are embedded.

Emotion recognition
Cultural awareness
Naturalistic databases
Recommender systems
UX, IxD, HCI