Mathematicians offered literary scholars a new way of determining the authorship of texts
Scientists from the Mathematics and Mechanics Department of St. Petersburg State University and Computer Faculty ORT Braude College (Israel) took part in the project. The mathematicians selected well known literary works for study: «Foundation», a sequence of seven science fiction novels by Isaac Azimov, «The Forsyte Saga» by John Galsworthy, «The Lord of the Rings», a three volumes novel by John Tolkien, and other works.
«To analyse the literary works we used the same principles of working with big data as for the analysis of changes in air temperature over a long period of time or the chemical composition of ice cores from a four-kilometre well in the central region of Antarctica in order to estimate climate change on Earth during thousands of years,» - said Oleg Granichin, Doctor of Physical and Mathematical Sciences, Professor at St. Petersburg State University and one of the authors of the study.
The scientists took part of text about the size of a page as a unit of information. As a result they got graphs describing the author’s style. Dr Granichin said that the peculiarities of a lexicon, frequency of certain turns of speech, the author’s intonation and rhythm were not taken into account in the study. «Literary scholars have used these characteristics before, without turning to big data analysis methods,» he said.
The resulting algorithm, said Dr Granichin, might become a useful tool for literary critics. It is suitable, for example, for finding out whether one or several authors created a certain literary work.