Čech Radek: Tematická koncentrace textu v češtině
The purpose of this book is to present a systematic analysis of a method to measure a thematic text property, termed thematic concentration, and to introduce ways of applying this method in textology. The method is based on frequency characteristics of a text. Select properties of rank frequency distribution of words are used to detect thematic words, i.e. words representing central topics of the text. Moreover, the method allows to quantify the thematic weight of these words and, consequently, to quantify a degree of the thematic concentration of the whole text. Differences between the thematic concentrations of particular texts (or groups of texts) can be statistically tested. Further, this book studies relationships between the thematic concentration and the vocabulary richness, as well as between the thematic concentration and the keyword analysis. The final part of the book is devoted to the application of this method in textology for analysis of the associated structure of a text and for classification of texts. As for the former, the method allows detection of statistically significant associations among thematic words in a text. Regarding the latter, particular registers such as fiction, scientific texts, journalistic texts, etc. differ significantly with regard to the thematic concentration. The method can also be used for the analysis of authorship.