A study carried out by researchers at the University of Barcelona and the Universitat Rovira i Virgili has enabled the development of the Spanish Media Monitor ( SMM ) project, a free, publicly accessible platform that analyses how the nightly news TV programmes of Spain's leading television channels cover current political affairs. Using artificial intelligence and data analysis techniques, the tool identifies the news items covered by each channel, calculates the airtime devoted to each one, and analyses mentions of political parties and the main political leaders, as well as the tone of the coverage.
The research and development were carried out by Luis Menéndez, from the UB's Department of Economics and Manuel Lleonart, from the URV's Department of Economics. The project aims to bring academic research closer to the public and foster a more critical and informed understanding of the media. It also seeks to create a database that allows the study of the evolution of media coverage over time, especially during election periods, and to provide objective and systematic data for professionals and researchers.
The SMM project uses AI to analyse the news programmes of Spain's five main television channels - La 1, Antena 3, Cuatro, Telecinco and laSexta. It first obtains a detailed transcription of the news broadcasts, and then a large language model (LLM) divides the text into individual stories, determines their durations and generates a summary.
The artificial intelligence they use is trained to understand the message and identify indirect references. Then, the system compares the topics covered and the time devoted to each by two or more television channels.
Analysis of the agenda and the tone of the news coverage
One of the platform's indicators is the coverage divergence, which measures the extent to which channels distribute news airtime differently across the same topics. It is presented on a scale from 0 to 100: a value closer to 0 indicates similar news selection, whereas one closer to 100 reflects very different agendas. In this case, the indicator takes into account the topics covered in each news story and the airtime dedicated to them, but not journalistic neutrality.
