Remember the old adage “I don’t know art, but I know what I like”? Its 21st century version might be: “Art may not know me, but it knows what I like.”

That, at any rate, is one of the possible consequences of an experiment conducted in 2006 by computer science researchers at England’s University of Bath. The researchers coupled a Web cam to an electronic art display, using the camera to capture viewers’ facial expressions. Software then compared eight key elements of those facial expressions against a database of emotional responses and the digital art adjusted in kind. The picture’s colors and brushstrokes adapted to the viewer’s mood changes, so happy visitors saw lively art, whereas unhappy or angry ones were presented with darker and more somber pictures. (More information on the study.)

This experiment paralleled an ongoing Teradata project.

For years Teradata has worked closely with the University of Southern California’s Integrated Media Systems Center (IMSC) on the E-Motions Smart Machine Project, profiled in the second quarter 2002 issue of Teradata Magazine. By capturing the user’s posture and facial expressions and reviewing them against its emotional database, a self-service machine such as an ATM or a vending kiosk could vastly improve the personalization of the customer’s experience.

The customer relationship management (CRM) benefits are clear—and profound. Real-time tracking of emotional states during inquiries and transactions on an ATM, for example, are designed to not only improve the customer experience, but also create an effective picture for the vendor of what messages and strategies work and what ones don't.

Down the road, there are countless other applications that could utilize a system with these features. Imagine, for instance:

	Security systems that are constantly vigilant for the telltale emotional signals of impending threats.
	Training and educational systems that not only pace themselves to the student’s progress but also “understand” and respond to the student’s comfort, confidence and emotional state.
	Individualized attention and responses from our electronic devices, giving a new and deeper meaning to the term “personal computer.”

The development of systems that can respond—and, more importantly, adjust their response—to users’ emotional states remains one of the large goals of both theoretical computer science and practical business applications. A goal that grows closer to fruition every year.

Whether in clever but relatively simple applications such as the emotional art experiment at the University of Bath, or the ambitious and still-evolving E-Motions Smart Machines in Teradata/IMSC research, it’s clear that “E-Motions live!” T

Related information Computer art changes to suit mood, BBC News, August 5, 2006 Computer scientists have helped develop electronic artwork that changes to match the mood of the person who is looking at it. Using images collected through a Web cam, special software recognizes eight key facial features that characterize the emotional state of the viewer. It then adapts the colors and brush strokes of the digital artwork.

Teradata Magazine-December 2006