МАШИН ОРЧУУЛГЫН ЧАНАРЫН ҮНЭЛГЭЭНД ХИЙМЭЛ ОЮУНЫ АРГАЧЛАЛЫГ АШИГЛАХ БОЛОМЖ БА ҮР НӨЛӨӨ

(парафраз болон хураангуйлах хэрэглэгдэхүүн ашигласан загварчлалын жишээн дээр)

Abstract

This research examines the potential of Artificial Intelligence (AI) techniques and algorithms to improve and assess the quality of machine translation (MT) outputs, utilizing paraphrase and summarize model tools to enhance the results. In recent years, a multitude of studies have scrutinized the defaults and limitations of machine translation when compared to human translations, particularly within the realm of translation studies. These researches emphasize the efficiency of machine translation regarding time and cost, while also underscoring various quality-related issues. The significance of this research lies in its proposal to integrate paraphrasing and summarization techniques to elevate the quality and assessment of MT. AI models and algorithms, including GPT, Google Translate, Trados, DeepL, and Microsoft translation tools, have highlighted the necessity for a comprehensive evaluation of the quality discrepancies between machine-generated and human translations. Recent findings indicate that rather than outright rejecting or endorsing machine translation, it is essential to explore strategies for enhancing its accuracy. This article aims to advance translation quality and evaluation by incorporating paraphrasing and summarization tools. The integration of AI translation algorithms with MT systems has the potential to improve translation quality, optimize workflows, expedite results, and provide economic advantages in project execution. The results under this study indicate that the application of paraphrasing and summarization models can significantly enhance the quality of MT outputs, yielding more accurate and improved translations compared to original MT methods.