[ad_1]
When applied to datasets from the Long Beach area, the algorithms detected many more earthquakes, making it easier to find out how and where they started. And when applied to a data In the 2014 earthquake that also struck La Habra, California, the team observed four times more seismic detections in the “denoized” data than the number officially recorded.
This isn’t the only business applying artificial intelligence to earthquake hunting. Researchers from Penn State are training deep learning algorithms to accurately predict how changes in measurements might be. to indicate future earthquakes—a task that has puzzled experts for centuries. And members of the Stanford team have previously trained models for phase selection or measuring the arrival times of seismic waves within an earthquake signal, which can be used to predict the location of an earthquake.
Deep learning algorithms are particularly useful for earthquake monitoring because they can take the burden of human seismologists, said Paula Koelemeijer, a seismologist at Royal Holloway University in London who was not involved in this study.
In the past, seismologists would look at graphs produced by sensors that recorded the motion of the ground during an earthquake and visually identify the patterns. Koelemeijer says deep learning can help cut through large volumes of data, making this process faster and more accurate.
“It shows [the algorithm] “Working in a noisy urban environment is very rewarding, because dealing with noise in urban environments can be a nightmare and very challenging.”
[ad_2]
Source link