One Concern is an AI company (人工智能公司) in California. It is using machine learning and AI to help firefighters. The company hopes its software (软件) can make exact predictions about earthquake damage. It can tell firefighters how to plan for an earthquake and how to do their jobs better when it happens.
Nicole Hu, a technology officer from One Concern, says the key is to feed the computers a lot of data (数据) about an area. The data may include information about homes and other buildings, the natural environment, and the size of an earthquake. The computers then use the information to make predictions about what would happen if there were an earthquake in the area. It then uses data from past earthquakes to see whether its predictions are any good, and changes its predictive models accordingly.
Gregory Deierlein, earthquake engineer at the Stanford University, says one of the most surprising things about the company's software is its ability to make use of data from an earthquake as it's happening, and to change its predictions in real time.
In the past, researchers would collect data after an event and a few years later they'd produce new models, says Deierlein. Now the new models appear in a matter of minutes.
Dan Ghiorso leads the Woodside Fire Protection District (区) near San Francisco, which covers about 32 square miles. He says in the past, when an earthquake hit, he'd have to make educated guesses (based on experience and training) about what parts of this district might be badly damaged and then drive to each place to have a look around. He hopes One Concerns' software will change that. "Instead of driving thirty-two square miles, in fifteen minutes on a computer I can get a good idea of the damage," he says. ‘‘Instead of me taking my educated guess, they're putting science behind the predictive models."
