Researchers use machine learning for the first time to predict timing of earthquakes

Researchers from the US and the UK have employed machine learning techniques to predict earthquakes.

As per a research published in AGU journal - Geophysical Research Letters, conducted by University of Cambridge, Los Alamos National Laboratory and Boston University researchers in an experiment that was performed in the laboratory setting identified hidden signals that lead up to earthquakes, and used this 'fingerprint' to train a machine learning algorithm to predict future earthquakes.

The researchers used steel blocks to closely mimic the physical forces at work in a real earthquake and also recorded the seismic signals and sounds that were emitted. Machine learning was then used to find the relationship between the acoustic signal coming from the fault and how close it is to failing.

"This is the first time that machine learning has been used to analyse acoustic data to predict when an earthquake will occur, long before it does, so that plenty of warning time can be given - it's incredible what machine learning can do,"  Professor Sir Colin Humphreys of Cambridge University was quoted saying in the report by Eurek Alert.

The development is indeed a great leap in attempts to predict natural calamities. Though it is not hard to pinpoint earthquake-prone areas, as this could be done by identifying areas that had suffered the calamity. However, what remains a big restriction is the inability of scientists to predict the time when the calamity will occur.

The result that is derived from the experiment can not only be applied to predict earthquakes but also other natural calamities like avalanches, landslides among others.