Diagnosing editorial strategies of Chilean media on Twitter using an automatic news classifier

Abstract

In Chile, there is no independent entity that publishes quantitative or qualitative studies that can provide the tools to understand how the traditional media environment has adapted to the social web. Nowadays, Chilean newsreaders are increasingly using social networks as their primary source of information. In this regard, Twitter plays a central role as it is considered, among the users, as the most influential news source on social networks; consequently, mainstream media are making efforts to develop different strategies to increase their audience and influence on this platform. Nevertheless, it is possible to affirm that there is a lack of tools that can serve to analyze these strategies. The following article intends to propose a methodology based on data mining techniques to provide the tools to carry out the analysis of the new Chilean media environment. Crawling techniques were used to mine news feeds from 37 different Chilean media that are currently active on Twitter; moreover, to provide several indicators to compare them. Thus, the volumes of production were analyzed in terms of their potential audience and NLP techniques were used to explore the contents of production, their publishing standards, and their geographical coverage.