This study describes the experimentally developed technique of sentiment analysis in an automatic semantic parser that processes daily news input blocks. The 2022 time period was used to analyze news in Telegram channels published by twelve official news agencies and five influential non-official news sources. In the advanced analytical system that we present, automatic learning is combined with a frame-based approach to identify patterns that influence judgments based on a frame representation of the text. The parser analyzes the syntactic structure of each sentence input, constructs its semantic representation, and associates it with one of 3800 semantic frames extracted from a large text corpus. The research revealed that the relative frequency and combination of frames varied between the two groups of news sources. The results prove that news is conceptualized and represented by various media sources in very different ways with the potential to affect the recipient's beliefs and emotions.

The study presents the experimentally developed technique to determine how and where semantic categories are represented in the human brain. For this purpose, the combination of linguistic annotation of oral texts in Russian with the registration of BOLD signal in functional MRI subjects’ testing was applied. The goal is to correlate groups of words in input texts and patterns of brain activation. The algorithm of the texts’ markup consisted of four sequential procedures: (1) time annotation, (2) lemmatization, (3) vectorization, (4) indexation of words-attributes. Regarding the issue of brain activity allocation, the corresponding activation distributed nature can be noted. It proves our hypothesis that conceptual information during text comprehension is processed in different (however specific) areas of the brain and is not compactly localized. The main result is that the allocated lexical clusters are composed of paradigmatically close words (synonyms, hyponyms) and situationally related words.

F2 syntactic parser is a part of F2 emotional robot, designed to support natural emotional communication with the helpof gestures, facial expressions and speech. The parser constructs syntactic and semantic representations (frame networks)of an input text, saves them to memory (database) and selects a communicative reaction for the robot in BML (behaviormarkup language) format. The model of reactions and inferences is based on scripts if-then operators, competing for theprocessing of semantics. In particular, scripts detect emotionally relevant meanings: when it is declared, that somebodythreatens the robot, does not care about it, behaves inadequately 13 negative scripts, and also when the robot is superior,attracts attention, etc 21 positive scripts. Parser may run in a standalone mode, daily processing sentences from news andblogs. Balancing of scripts allows us to tune the understanding and reproduce different emotional profiles for the robot.(Research is supported by RSF, project No 17-78-30029).

We design F-2 companion robot, supporting natural multimodal communication. The robot is operated by a set of scripts,triggered by input speech and generating behavioral patterns in BML format. To make robots behavior as close as possibleto humans, we extract natural communication patterns from the Russian Emotional Corpus REC (over 400.000 annota-tions), reproduce key patterns in Blender 3D editor and export them to MySQL database (n = 220). For each generatedBML the software retrieves the corresponding movement from the database, joins compatible patterns and performs themon the robot. Robot can also receive the coordinates of surrounding human faces and simulate direct gazes towards theeyes of the addressee. It can also perform oriented (pointing) gestures: switch between directions or between severalinterlocutors. This allows us to model complex robot behavior, as shown in our experiment, increasing human satisfactionfrom robot-to-human interaction (Research is supported by the Russian Science Foundation, project No 19-18-00547).

Aggression speech patterns (ASP) strongly influence modern culture and ideology, they are regular source domain forconceptual metaphors. The study was based on THREAT-corpus (Russian language, 5 mln words) which was constructedto study ASP and contains fiction, non-fiction, news texts. The aim of the study is to investigate non-metaphoric andmetaphoric ASP in Russian. A semantic parser was designed to automatically process texts and construct conceptualrepresentations: They killed all the enemies (non-metaphoric)/He killed the time (metaphoric) [Ag-CAUSE HARM-Pat]. After extracting conceptual representations the parser evaluates them as aggressive or non-aggressive. An exampleMechanical toys pushed forward the imagination of scientists is evaluated as aggressive. Although this evaluation is false-positive, it reveals the conceptual metaphor where mental causation is described as voluntary action. This set of methodsmakes possible to collect, detect and describe ASP in diverse types of discourse and, consequently, to analyze the cognitivenature of aggression.

The results were obtained in the course of a two-stage study. At the first stage (2018) linguists analyzed the conceptualdomain sociocultural challenges on the basis of purposely elaborated Russian language THREAT-corpus (10.4 m words)and built a frame of the domain. At the second stage (2018-2019) the research was carried out with methods of automatedtopic modeling for two Russian language corpora: THREAT-corpus and alternative corpus collected using WebBootCaTtool in the SketchEngine corpus management system. Methods of topic modeling (PLSA, LDA, BigARTM et al.) allowedeliciting thematic profiles for texts of both corpora. Comparison of two datasets was carried out by applying set theory,graph theory, and probabilistic analysis. Combining topic modeling with linguistic frame analysis resulted in more pre-cise configurations of cognitive models in the conceptual domain sociocultural challenges. Word frequency for lexemesmanifesting sociocultural challenges proved to be an important factor of conceptual structures representation.