Automatic Extraction of Aggression Speech Patterns in the THREAT-corpus
- Vera Zabotkina, Center for Cognitive Programs and Technologies, Russian State University for the Humanities, Moscow, Russian Federation
- Boris M. Velichkovsky, Center of NBICS-Technologies, Kurchatov Institute, Moscow, Russian Federation
- ELENA POZDNYAKOVA, Center for Cognitive Programs and Technologies, Russian State Univercity for the Humanities, Moscow, Russian Federation
- Dmitry Orlov, Center for Cognitive Programs and Technologies, Russian State Univercity for the Humanities, Moscow, Russian Federation
- Artemy Kotov, Center of NBICS-Technologies, Kurchatov Institute, Moscow, Russian Federation
AbstractAggression speech patterns (ASP) strongly influence modern culture and ideology, they are regular source domain for conceptual metaphors. The study was based on THREAT-corpus (Russian language, 5 mln words) which was constructed to study ASP and contains fiction, non-fiction, news texts. The aim of the study is to investigate non-metaphoric and metaphoric ASP in Russian. A semantic parser was designed to automatically process texts and construct conceptual representations: 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 example Mechanical 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 methods makes possible to collect, detect and describe ASP in diverse types of discourse and, consequently, to analyze the cognitive nature of aggression.