THE METHOD OF IDENTIFYING INDIRECT SIGNS OF CORRUPTION ACTS BASED ON VIDEO RECORDINGS OF SPEECHES OF CIVIL SERVANTS

Abstract

At the moment, the problem of countering corruption in the field of public service still does not lose its relevance. It is assumed that when committing acts of corruption, people show certain verbal and non-verbal signals, with the help of which it is possible to identify signs of corruption. The article deals with the problem of violation of anti-corruption legislation in state and municipal institutions from the point of view of the psycho-emotional state of officials. It is proposed to use machine learning methods to analyze video and audio recordings of speeches with unprepared speech of officials of various levels of government, where they answer questions from journalists and the public, in order to determine emotions and identify aggression, uncertainty, and evasiveness in answers. The results of the study may be useful for government agencies involved in the fight against corruption, as well as for the public interested in transparency and honesty of the activities of civil servants.