Spectral Analysis Applications to the Literary Texts Study
This paper presents an approach for dynamic modeling of writing process. A text is divided into sequence of sub-texts that are described through occurrences of character $N$-grams. The Mean Dependence similarity measures the association between a present sub-text and a number of preceding ones and transforms a text into a time series, which is supposed to be weak stationary if the text is created using single writing style. A periodogram of this signal estimates its Power Spectral Density providing a spectral attribute of the style. Numerical experiments demonstrate high ability of the proposed method in authorship identification and the revealing of writing style evolution.