Дослідження методів оцінки сентименту діалогових повідомлень

Abstract

In this theses proposes the task of text sentiment analysis or sentiment analysis involves determining the emotional attitude of the author towards a specific object described in the text. This task is one of the most relevant tasks in natural language processing (NLP). Sentiment analysis is used to evaluate the quality of goods and services based on Internet user reviews, to detect criminally significant content, to determine the authorship of texts, to forecast various economic indicators, and to generate texts with predetermined emotional coloring. The amount of information in electronic form is growing exponentially. Therefore, manual analysis is impossible, leading to the need for automatic methods and tools for analyzing textual information, including methods and tools for automated sentiment analysis.