Exploring lexical diversity in Basque news: original vs. machine-translated texts

The rapid advancement of machine translation (MT) has led to increased exposure to MT-generated content. Given the potential influence of this technology on language use, this study compares the lexical diversity of machine translated and originally-pro­duced online news articles in the Basque language. Combining established automatic metrics with detailed manual analysis, we conduct a comprehensive evaluation of multiple dimensions of lexical diversity. Our findings indicate a striking similarity in lexical richness between the two text types, with few aspects in which original texts surpass the lexical diversity of machine-translated texts.