Journal of Earth Science & Climatic Change
Open Access

Abstract

T his paper is shown that some adapted preprocessing can help neural networks in classifying climatic change shapes by using broad-band seismograms. The extracted features by statistical and neural methods show the shape discrimination problem. The extracted spectral curve by a three-layered perceptron from Long-Period records on seismograms gives the best recognition rates among many classical neural and non neural discrimination methods. A preliminary experiment with computer simulation showed that this approach is promising the recognition and segmentation of characters on earthquake records can be successful to predict the regional climatic changes.

Biography

Mostafa Allameh Zadeh has completed his Ph.D. from International Institute of Earthquake Engineering and (IIEES). He is the director of CTBTO Seismic Networks in Iran, a premier Bio-Soft service organization. He has published more than 25 papers in Seismology journals and serving as an editorial board member of repute.