| Abstract [eng] |
The author of each literary text has individual character traits, habits, interests and etc. Obviously, in some way it should be reflected in their works. This thesis analyzes the first six samples of every novel from the twelve novels set "Pora vienam vakarui". The aim of the work is to determine whether the author of the literature text can identify the lingvo-statistical methods. The tasks are: to digitize and sort out linguistic data; elect the sample and identify the linguistic characteristics; determining whether the linguistic data under consideration is independent and incidental; evaluate the data normality by checking the corresponding statistical hypotheses; determining the numerical characteristics of the number of letters and the number of words in sentences by literature texts; to check the hypothesis of averaging equality using empirical averages of the number of letters and words in the sentences; to perform a correlation analysis between the words in the sentences and the number of letters in the sample; to determine statistical frequencies of the number of letters of the Lithuanian alphabet, later, to use them to verify the hypotheses about equality of binomial distribution parameters. Following a lingvo – statistical analysis, it has been found that: linguistic data with high reliability is random and independent, and normally distributed; linguistic characteristics of the number of letters and words in the sentence are not very constant, moving in different works of the author, and therefore can not be regarded as the identity of the author of literature text; the correlation between the number of letters and the number of words is similar and statistically significant, therefore this characteristic may be appropriate for identifying the author of the literature text; the frequencies of the letters of the Lithuanian alphabet are very similar to the analyzed texts, and therefore can be used as a measure of the identity of the author of a fiction literature text. |
