Recycling a genre for news automation

Abstract The amount of available digital data is increasing at a tremendous rate. These data, however, are of limited use unless converted into a user-friendly form. We took on this task and built a natural language generation (NLG) driven system that generates journalistic news stories about elections without human intervention. In this paper, after presenting an overview of state-of-the-art technologies in NLG, we explain systematically how we identified and then recontextualized the determinant aspects of the genre of an online news story in the algorithm of our NLG software. In the discussion, we introduce the key results of a user test we carried out and some improvements that these results suggest. Then, after relating the news items that our NLG system generates to general aspects of genres and their evolution, we conclude by questioning the idea that journalistic NLG systems should mimic journalism written by humans. Instead, we suggest that developmental work in the field of news automation should aim to create a new genre based on the inherent strengths of NLG. Finally, we present a few suggestions as to what this genre could include.