Automated Writing of COVID-19 News

This case study shows how the publisher news.de uses AX Semantics and DataCater to automate the writing of COVID-19 reports for more than 400 regions in Germany.

news.de aims to publish current and high-quality reports on the latest development of the COVID-19 pandemic. Reports should provide quick and accurate access to information, such as the number of active cases, the availability of intensive-care beds, or the progress in vaccination. news.de does not only want to publish such reports for entire countries, as done by most news providers, but also on a state and county level, since the current state of the COVID-19 pandemic differs strongly from region to region. Given that Germany has more than 400 counties, manually writing such reports would make it impossible to publish them daily.

news.de uses the AX Semantics NLG Cloud and the DataCater platform to automate the production of reports on the current situation of the COVID-19 pandemic for more than 400 different geographical regions. Reports are generated at least twice per day and are based on data provided by the Robert Koch Institute.

As data is the base of automated story writing with AX Semantics, there is always the challenge to get enough, structured and accurate data. That's why news.de uses DataCater to build the data pipelines, which extract data from various APIs of the Robert Koch Institute, prepare the data for the text production, and send the prepared data to the AX Semantics NLG Cloud. The AX Semantics NLG Cloud uses the data to automatically produce optimized texts, which are sent to the content management system of news.de.

By using AX Semantics for automating the production of COVID-19 reports, news.de was able to serve more than 25,000 article updates and reach a local audience of more than 800.000 new visitors per month.

Using DataCater for building and managing data pipelines allows the data engineering team of news.de to save approximately 40 % of their time, otherwise spent on manually crafting code, and increase the robustness of the data preparation processes.


“Traditionally being a publisher focussed on national and international headlines, news.de was able to implement its first large-scale local news coverage thanks to the data-handling infrastructure provided by DataCater. Encouraged by the achievements, we will further expand upon the capabilities of DataCater, AX Semantics and its potential for content production at scale on a broader range of topics.”

André Tschachtli (CTO, news.de)

news.de

news.de is a cross-regional and independent news platform from Germany. news.de publishes a broad range of articles each day, covering the latest news in politics, economy, and sport, and has more than 9 million unique visitors each month.

AX Semantics

AX Semantics is an AI-powered, Natural Language Generation software company with its roots grounded in content and storytelling. AX Semantics builds a sophisticated, yet easy to use SaaS-based software, which makes automated content generation accessible to customers of all sizes and is used widely within the e-commerce, business, finance, and media publishing sectors.

DataCater

DataCater is the low-code platform for streaming data pipelines, which makes Apache Kafka®-powered streaming data pipelines accessible to data teams. DataCater can be used to integrate a wide range of data systems, such as database systems and web APIs, in real-time.