Affective computing, also known as AC or emotion AI, is an area of study within cognitive computing and artificial intelligence that is concerned with gathering data from faces, voices and body language to measure human emotion. An important business goal of AC is to build human-computer interfaces that can detect and appropriately respond to an end user's state of mind.
Affective computing has the potential to humanize digital interactions and offer benefits in an almost limitless range of applications. For example, in an e-learning situation, an AC program could detect when a student is frustrated and offer expanded explanations or additional information. In telemedicine, AC programming can help physicians quickly understand a remote patient's mood or look for signs of depression. Other business applications currently being explored include customer relationship management (CRM), human resource management (HRM), marketing and entertainment.
A computing device with emotion AI programming gathers cues about a user's emotional state from a variety of sources, including facial expressions, muscle tension, posture, hand and shoulder gestures, speech patterns, heart rate, pupil dilation and body temperature. The technology that supports emotion measurement and analysis includes sensors, cameras, big data, deep learning analytics engines. As of this writing, RESTful APIs are available to measure human emotion from companies such as Affectiva, Humanyze, CrowdEmotion and Emotient. IBM Watson APIs include Tone Analyzer and Emotion Analysis.
The term affective computing is generally credited to Rosalind Picard, a computer scientist at MIT and founder of Affectiva. In psychology, the word affect is used to describe a patient's emotional tone.
Continue Reading About affective computing (emotion AI)
- MIT Media Laboratory's Affective Computing pages offers a list of readings and more information about affective computing.