We introduce a class of non-stationary processes, or signals, that are close to stationary and call them locally stationary. They arise in many applications in seismology, in speech analysis and elsewhere. We show that their local spectral characteristics can be obtained eeciently using an adaptive windowed Fourier transform. We illustrate with some examples from seismology the use of our metho...

Machine learning (ML) is a collection of methods used to develop understanding and predictive capability by relationships embedded in data. ML are becoming the dominant approaches for many tasks seismology. data mining techniques can significantly improve our seismic processing. In this review we provide comprehensive overview applications earthquake seismology, discuss progress challenges, off...