From entry Fourier transform in glossary Artificial Intelligence: humans at the heart of algorithms
A Fourier transform converts time series or other sequential data into a a combination of sine and cosine waves of different frequencies. This is similar to the frequency level indicators that are seen in high-end amplifiers, but at a much finer scale. Typically the signal is split into frequencies using some power of 2, as this can be calculated using a Fast Fourier Transform, but in principle any range of consecutive frequencies can be used.
Used in Chap. 3: page 43