WebNLG Challenge: Human Evaluation Results
Anastasia Shimorina, Claire Gardent, Shashi Narayan, Laura Perez-Beltrachini · HAL (Le Centre pour la Communication Scientifique Directe) · 2018
This report presents the human evaluation results for the WebNLG Challenge which was held in 2017. The automatic evaluation results can be foundin [Gardent et al., 2017a]. In this report, we describe human evaluation design, communicate the results, and explore correlation between automaticand human assessments.