With software automatically producing texts in natural language from structured data, the evolution of natural language generation (NLG) is changing traditional news production. The paper first addresses the question whether NLG is able to perform the functions of professional journalism on a technical level. A technological potential analysis therefore uncovers the technological limitations and possibilities of NLG, accompanied by an institutional classification following Weischenberg, Malik, and Scholl. Overall, NLG is explained within the framework of algorithmic selection and along its technological functionality. The second part of the paper focuses on the economic potential of NLG in journalism as well as indicating its institutionalization on an organizational level. Thirteen semi-structured interviews with representatives of the most relevant service providers detail the current market situation. Following Heuss, the development of the NLG market is classified into phases. In summary, although the market for NLG in journalism is still at an early stage of market expansion, with only a few providers and journalistic products available, NLG is able to perform tasks of professional journalism at a technical level. The analysis therefore sets the basis to analyze upcoming challenges for journalism research at the intersection of technology and big data.
Konstantin Dörr examines Natural Language Generation (NLG) and its uses for preparing news. He includes analysis based on talks with 13 service providers.
- Konstantin Nicholas Dörr, University of Zurich, Institute of Mass Communication and Media Research, Media Change & Innovation Division, Zurich, Switzerland.
Dörr, K. (2016) ‘Mapping the field of Algorithmic Journalism,’ Digital Journalism, 4(6):700-722. DOI: https://doi.org/10.1080/21670811.2015.1096748