Giving computers capacity to learn.

Who says you can't find good help these days? Ray Eberts, associate professor of industrial engineering, Purdue University, predicts that computers driven by neural networks soon may perform such tasks as sorting the junk out of electronic mail, fetching interesting stories from newspapers on the information superhighway, or telling drivers on an actual highway that they really will like the restaurant that's just over the next hill.

A neural network is a computer program, modeled roughly after the brain, that can learn to perform tasks and make decisions based on past examples or experiences, Eberts explains. "Any decision requiring expertise, such as medical diagnosis, legal decisions, or financial investments, could be assisted by using neural nets. Just as people have procedures for learning from experience, the neural network also has learning procedures stored in the program. Also like people, neural nets can recognize patterns and analyze subjective information such as speech, fingerprints, or stock performance better than standard programs, which are designed primarily for very precise calculations. Neural nets also can learn to give advice on how to perform tasks more efficiently."

They "learn" by gathering and storing information from the computer...