..., the EA measure currently relies on human rating and is thus time-consuming, making the use of large language models (LLMs) an attractive alternative. It is essential to consider two significant constraints when implementing LLMs as a solution: the choice of LLM and the impact of domain-specific datasets...