The work is devoted to the identification of links between the automatically distinguished features of the text and the autor's psychological characteristics. On the basis of the analysis of works carried out in Russia and in the world, and on the results of our study, it is proposed to consider the system of automatic text analysis as research tools for a work of a psychologist with large text corpora. The results of a study conducted using the linguistic analyzer PLATIn, developed on the basis of the processor Exactus Expert in ISA RAS, are presented. The study included psycholinguistic and lexical-frequency analysis of texts created by the subjects (142 people, students and adults, Moscow and Kurgan), who underwent psychodiagnostic examination. The lexical analysis was based on the specially designed dictionaries of emotional topics (about 53 thousand lexical units). The analysis of correlations between the text parameters (12 psycholinguistic and 30 lexical) and the psychodiagnostic parameters (81 scales of 10 questionnaires) was carried out. The study showed the sensitivity of the lexical and psycholinguistic parameters to some psychological characteristics. The conclusion about the expediency of the automatic text analysis in psychological population studies is made.
DOI: https://doi.org/10.14357/20790279190308
PDF at the Institute for Systems Analysis website (in Russian): www.isa.ru/proceedings/images/documents/2019-69-3/91-99.pdf
Page at the Proceeding of the Institute for Systems Analysis of the Russian Academy of Science website: http://www.isa.ru/proceedings/index.php?option=com_content&view=article&id=1097&Itemid=&lang=en
S. N. Enikolopov, Y. M. Kuznetsova, A. N. Minin, M. Y. Penkina, I. V. Smirnov, M. A. Stankevich, N. V. Chudova. Text features and psychological characteristics: an empirical study of computer // Works of the Institute of Systems Analysis of the Russian Academy of Sciences. - 2019. - Vol. 69. - №. 3. – Page 91-99.