СЕМАНТИКО-СТАТИСТИЧЕСКИЙ АЛГОРИТМ ОПРЕДЕЛЕНИЯ КАТЕГОРИЙ АСПЕКТОВ В ЗАДАЧАХ СЕНТИМЕНТ-АНАЛИЗА

In the modern world, one of the important communication channels is the Internet. Trade,promotion of services is carried out through electronic platforms. Social networks and instantmessengers are becoming the most important communication channel and a powerful tool forinfluencing public opinion. A significant amount in all published content falls on texts written innatural language. Therefore, the problems of natural language processing (NLP) and naturallanguage understanding (NLU) today are one of the key ones. Under the influence of commercialinterests, the field of automatic aspect-based sentiment analysis is actively developing. This tasksignificantly depends on specific subject areas, and therefore the issue of quick and effective adaptationof existing models to new domains is very acute. The paper proposes a hybrid method ofaspect-oriented analysis, based on data extracted from common dictionaries and domain-orientedtexts. The novel method for constructing a condensed semantic graph based on unstructured domain-dependent texts is proposed. Numerical metrics to assess the significance of individual termswithin the entire domain are introduced. An algorithm for the text categorization based on theselection of semantic clusters within a condensed domain-specific graph is proposed. A method forassessing the sentiment of domain-oriented texts based on statistical data, including the joint useof a tone lexicon and a condensed domain-specialized graph, is proposed. The results of experimentsare presented, allowing for evaluation of the quality of the algorithms.