Dataset for Quantitative Analysis of NMT, LLM, and Human Translations in Narrative Medicine

This dataset contains quantitative linguistic measurements derived from machine-translated and human-translated narrative medicine texts. The corpus includes outputs produced by neural machine translation (NMT) systems, large language models (LLMs), and professional human translators.The dataset was constructed to support comparative quantitative analysis of translation outputs, focusing on lexical diversity, syntactic complexity, discourse-level cohesion, and pragmatic features. Linguistic indicators were extracted using automated computational tools. All data were preprocessed prior to analysis. The dataset is intended for research use in quantitative linguistics, translation studies, and computational linguistics, and may be reused for replication studies or further methodological development.