236 | nº 29, pp. 235-254 | doxa.comunicación

July-December of 2019

Automated sports journalism. The AnaFut case study, the bot developed by El Confidencial...

ISSN: 1696-019X / e-ISSN: 2386-3978

1. Introduction: automated journalism and newsroom automation

The relationship between information, journalists, newsrooms, media, algorithms, bots and Artificial Intelligence has been defined in various ways by many authors in recent years. They all have one common denominator, which is the undeniable relationship between the worlds of “computing, social science and communication” (Flew et al., 2012). For Carlson (2014), these are “algorithmic processes that convert data into informative, narrative texts with little or no human intervention other than initial programming”. Graefe (2016), on the other hand, defines the phenomenon as a “process of using software or algorithms to generate news automatically without human intervention after the initial programming of the algorithm”.

Thus, on the basis of these definitions, we also encounter different nomenclatures, ranging from automated journalism (Graefe, 2016), to algorithmic journalism (Dörr, 2016), to robotic journalism (Oremus, 2015). All of these are the result of a technological process that began at the end of the 1980s of the twentieth century (Túñez López, Toural Bran, Cacheiro Requeijo, 2018), which focused on the production of journalistic content, analysis and visualisation (Carlson, 2014; Gao, Hullman, Adar, Hecht and Diakopoulos, 2014; Diakopoulos, 2014; Young and Hermida, 2014; Broussard, 2015; Cervera, 2017).

In addition to the tasks resulting from the definition itself, for some authors (Lokot and Diakopoulos, 2016), the main tasks of bots have to do with participating in dissemination on social platforms, supporting and managing broadcasts, and adding web content (Starbird, Leysia, Hughes and Vieweg, 2010; Mittal and Kumaraguru, 2014), while at the same time being capable of identifying facts and events of journalistic relevance for later difussion (Steiner, 2014).

1.1. State of the issue in the academic field

In the academic field, studies on the application of Artificial Intelligence for the automated creation of news have increased in the last 15 years both in the quantity and quality of the research, and in the number of deliberations on the phenomenon (Slater y Rouner, 2002; Powers, 2012; Levy, 2012;Karlsen y Stavelin, 2013; Matsumoto, Nakayama, Harada y Kuniyoshi, 2007; Napoli, 2012; Van Dalen, 2012; Clerwall, 2014; Edge, 2014; Latar, 2014; Carlson, 2014; Oremus, 2015;