Abstract
The present paper investigates the characteristics of computer-assisted translation tools from both theoretical and practical perspectives. It addresses current issues related to the development of new typologies and classifications that may be useful in modern computer-assisted translation training and in solving applied tasks arising during the development of digital translation products. Frequency-based errors occurring in machine translation serve as the basis for this study. The authors present the results of a comparative analysis of texts translated into English using digital translation tools. The classification of transformation-induced translation errors is based on three types of translation transformations: lexical, grammatical, and lexico-grammatical. The analysis identified eight subgroups of frequent machine translation errors within these categories. Based on the extralinguistic analysis, the study proposes the main factors contributing to the occurrence of such errors. The findings indicate that machine translation errors may be associated with the use of computer-assisted translation tools without adherence to established standards, particularly regarding variations in the transliteration of proper names in Kazakhstani scientific literature. The study contributes to a better understanding of the limitations and challenges of computer-assisted translation tools and provides recommendations for their effective use in both translation practice and research activities.