Natural Language News Generation from Big Data

In this paper, we introduce an NLG application for the automatic creation of ready-to-publish texts from big data. The resulting fully automatic generated news stories have a high resemblance to the style in which the human writer would draw up such a story. Topics include soccer games, stock exchange market reports, and weather forecasts. Each generated text is unique. Readyto-publish stories written by a computer application can help humans to quickly grasp the outcomes of big data analyses, save timeconsuming pre-formulations for journalists and cater to rather small audiences by offering stories that would otherwise not exist. 

• Bastian Haarmann
• Lukas Sikorski

• Big data
• natural language generation
• publishing
• robotic journalism.

[1] N. Bouayad-Agha, G. Casamayor & L. Wanner: Content selection from
an ontology-based knowledge base for the generation of football
summaries. ENLG 2011 Proceedings, pp. 72-81. Nancy, FR. 2011.
[2] N. Effingham: An Introduction to Ontology. Wiley-Blackwell: Hoboken,
NJ, USA. 2013.
[3] T. Hoppe: Messung des Nutzens semantischer Suche. In: B. Humm, A.
Reibold & B. Ege (ed.): Corporate Semantic Web. Berlin: Springer.
2015.
[4] E. Reiter & R. Dale: Building Natural Language Generation Systems.
Cambridge University Press. Cambridge, UK. 2000.
[5] M. Bollmann: Adapting SimpleNLG to German. Proceedings of the 13th
European Workshop on Natural Language Generation (ENLG), pp. 77-
81. Nancy, FR. 2011.
[6] A. Gatt & E. Reiter: SimpleNLG: A realization engine for practical
application. Proceedings of the 12th European Workshop on Natural
Language Generation (ENLG), pp. 72-81. Athens, GR. 2009.
[7] N. Chomsky: Syntactic Structures. Mouton. Den Haag, NL. 1957.
[8] N. Chomsky: Lectures on Government and Binding. Mouton de Gruyter.
Berlin, DE. 1993.
[9] L. Tesnière: Eléments de syntaxe structurale. Klincksieck. Paris, FR.
1959.