Specifics and adaptation of machine learning in modern humanities amid turbulence, uncertainty and systemic risks: evidence from modern linguistics

Abstract

The article explores the specifics, features and directions of adapting machine learning technologies in modern humanities using the example of linguistics. The possibilities of ML for analyzing linguistic data are considered, as well as the problems of uncertainty, turbulence, systemic risks, algorithmic bias and interpretation of results. The need to transition to adaptive machine learning models that combine technological capabilities, expert humanitarian interpretation and a risk-oriented approach is substantiated.