Застосування машинного навчання в оцінці психоемоційних станів людини

Abstract

Mental disorders are a global problem, with statistics showing that more than 40% of people experience various forms of mental illness at some point in their lives. They cause a significant social and economic burden, accounting for 13.1% of the total number of years of life lived with a disability. The diagnosis of psycho-emotional conditions is often based on complex psychological tests. The experiment showed that during research, taking into account the main text with headings increases the overall accuracy of the models. Specifically, accuracy metrics are: logistic regression - 74.16%, BERT - 78.22%, LSTM - 69.36%.