My research focuses on earthquake engineering and seismology. My goals are twofold: (a) close the gap between earthquake engineering and geoscience (b) apply earthquake early warning systems (EEWS) to engineering structures. I have significant experience in engineering application of EEWS, real-time seismology, and engineering application of cognitive sciences. I mainly worked on solving advanced problems of real-time EEWS and I developed the second version ElarmS which is the core algorithm of the current running EEWS in California, USA. I had been evaluating and improving earthquake early warning algorithms in California. I solved problems which real-time operations encounter such as accurate estimation of magnitude/intensity and their relation between frequency and amplitude of ground motion, and improved association/alert criteria to reduce missed and false alerts.
My research interests moreover include the engineering adaptive systems of artificial intelligence. This secondary interest with respect to cognitive modeling focuses on the application of artificial neural network (ANN) in the seismological application including pattern recognition, self-organizing maps etc. My research demonstrates how to utilize advanced supervised learning ANN techniques for EEWS applications. I wrote an ANN-based general program in MATLAB. I have gained expertise in sophisticated artificial intelligence methods to tackle the problem in this seismological and earthquake engineering environment. I applied a couple of pattern recognition algorithms such as the mixture of Gaussian, k-means, Self-Organizing Maps, decision three, compared with linear and nonlinear discriminant functions to classy quarry blast explosions and earthquakes near Istanbul. I am recently working on convolution and sequential neural network modeling using Keras tool and TensorFlow in python language.