Emotionally Intelligent Information Systems
Emotionally Intelligent Systems are systems that are able to detect, process, interpret, and respond to human emotions in a digital form. Kurzweil (2005, p. 37) states that “the human ability to understand and respond appropriately to emotion (so-called emotional intelligence) is one of the forms of human intelligence that will be understood and mastered by future machine intelligence”.[1] Different steps are necessary for machines to be characterized as being emotionally intelligent.
Step 1: Detection of Emotions
Information systems need to be able to detect emotions. For doing so, several affect-detection channels have been suggested. Calvo and D'Mello (2010) list six affect detection channels emotionally intelligent systems can use:[2]
- Facial Expression
- Voice
- Body Language and posture
- Physiology
- Brain imaging and EEG
- Text
A different classification of affect-detection channels is to separate physiological (e.g.: facial expressions, body language and posture, brain imaging and EEG, eye movement) and non-physiological measures (e.g.: voice and text).
Affect-detection channels have also been combined. For instance, in a research setting, facial expressions have been combined with eye movement.[3]
Step 2: Processing of Data on Emotions
Depending on the affect detection channel, emotionally intelligent systems need to adopt different ways of processing and interpreting the produced data. Since data from affect-detection channels tends to be large in quantity and diverse in formats, big data analysis techniques are used to process such data. Software that helps processing such data is R and Python.
Step 3: Interpretation of Data on Emotions
Interpretation of data is strongly context-dependent. For instance, emotional facial expressions do not materialize themselves in the same way in all individuals. Or, changes in eye dilation can be an indication of an emotional reaction as well as cognitive load.[4][5] Such cognitive load reactions in pupils, in turn, are different dependent on individual characteristics.[6] Thus, similar to context knowledge that humans possess when interpreting emotions, emotionally intelligent systems need to acquire substantial context knowledge to interpret data in human emotions. Machine learning techniques can be applied to improve such interpretation.
Step 4: Responding to Emotions
Human emotion interacts with human cognition and sometimes overrides cognition.[7][8][9] Thus, emotionally intelligent systems that consider information on human emotions appropriately, have the potential to more effectively interact with humans. They can respond in an emotionally adequate way by, for instance, (1) displaying information at the right time depending on emotional states of humans, or (2) exhibiting emotions themselves.
References
- ↑ Kurzweil, R. (2005). The singularity is near: When humans transcend biology. New York (NY): Vikin. Search this book on
- ↑ Calvo, R., & D'Mello, S. "Affect detection: An interdisciplinary review of models, methods, and their applications". IEEE Transactions on Affective Computing. 1(1).CS1 maint: Multiple names: authors list (link)
- ↑ Fehrenbacher, D.D. "Affect Infusion and Detection through Faces in Computer-mediated Knowledge-sharing Decisions". Journal of the Association for Information Systems. 18(10).
- ↑ Hess, E.H. (1965). "Attitude and Pupil Size" (PDF). Scientific American.
- ↑ Kahneman, D. (1973). Attention and Effort. Prentice-Hall. Search this book on
- ↑ Hess, E.H., Polt, J.M. "Pupil size in relation to mental activity during simple problem-solving". Science. 143 (3611).CS1 maint: Multiple names: authors list (link)
- ↑ Forgas, J. (1995). "Mood and judgment: The affect infusion model". Psychological Bulletin. 117(1).
- ↑ Forgas, J., Eich, E. (2013). Affective influences on cognition: Mood congruence, mood dependence, and mood effects on processing strategies. New York: Handbook of psychology, Wiley.CS1 maint: Multiple names: authors list (link) Search this book on
- ↑ Cohen, J. D. (2005). "The vulcanization of the human brain: A neural perspective on interactions between cognition and emotion". The Journal of Economic Perspectives. 19((4).
This article "Emotionally Intelligent Information Systems" is from Wikipedia. The list of its authors can be seen in its historical and/or the page Edithistory:Emotionally Intelligent Information Systems. Articles copied from Draft Namespace on Wikipedia could be seen on the Draft Namespace of Wikipedia and not main one.
