Data mining application: neural networks on seismic data
Today, we continue our worldwide discovery of applications of data mining in a specific engineering domain. I’m sure you know a lot of applications of neural networks to face, speech or digit recognition. The paper by Huang et al. (1) is a good example of neural network’s application in an original area: diagnosis of a building using seismic response data. They present a procedure for diagnosing if a building has been damaged by earthquakes. A backpropagation neural network is used to compare buildings after a small earthquake (without damage) with data from damaged buildings (after big earthquakes).
(1) Huang C.S., Hung S.L., Wen C.M. and Tu T.T., A neural network approach for structural identification and diagnosis of a building from seismic response data, Earthquake engineering and structural dynamics, 32, 2003, pp 187-206.