Success prediction of online news about TV series with machine learning, Google Analytics, and Twitter

Aug 4, 2026

Digital journalism has adopted web analytics and trend analysis to evaluate how well online content performs. To optimize scarce editorial resources and increase visibility, this study introduces a cybermetric approach that uses machine learning to predict the success of online news covering television series, where virality is tightly linked to social networks. The method selects indicators and tools, collects data, applies multiple linear regressions, and validates prediction equations for accuracy.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages22

Reading Time55 min

AccessFree to download

Abstract

Introduction

Literature review

Digital journalism

Prediction of news success

Journal of Computational Social Science (2025) 8:78 [https://doi.org/10.1007/s42001-025-00412-9](https://doi.org/10.1007/s42001-025-00412-9) RESEARCH ARTICLE Success prediction of online news about TV series with machine learning, Google Analytics, and Twitter Víctor Yeste1,2 · Ángeles Calduch-Losa1 · José-Antonio Ontalba-Ruipérez1 · Jorge Serrano-Cobos1 Received: 10 October 2024 / Accepted: 3 July 2025 © The Author(s) 2025 Abstract Journalism has adapted to the digital environment using web analytics and trend analysis to measure the success of its content. To optimize resources and increase visibility, new information needs arise in the editorial process. Therefore, this study proposes a cybermetric methodology that employs machine learning to predict the success of online news about television series, a growing theme whose virality is closely related to social networks. The methodology design consists of selecting indicators and tools, data collection, multiple linear regressions to predict success indicators, and validating prediction equations to obtain their accuracy. Prediction equations of success indicators have been obtained using an online media outlet as a use case, segmenting the data into three sets: all articles, TV series articles, and trailer articles. Validation has allowed for the comparison of equations and the selection of the most accurate equation. This research provides a tool that can be integrated into the editorial process to optimize its strategy, and it is a starting point for future research to improve accuracy in multiple ways. Keywords Digital journalism · Digital marketing · Machine learning · Multiple linear regression · Social media analytics · Web analytics. 􀀍 Víctor Yeste vicyesmo@upv.es Ángeles Calduch-Losa [mcalduch@eio.upv.es](mcalduch@eio.upv.es) José-Antonio Ontalba-Ruipérez [joonrui@upv.es](joonrui@upv.es) Jorge Serrano-Cobos [jorserc2@har.upv.es](jorserc2@har.upv.es) 1 Universitat Politécnica de Valéncia, Valencia, Spain 2 Universidad Europea de Valencia, School of Science, Engineering and Design, Valencia, Spain 1 3 1 Introduction Digital journalism is, according to Nelson [38], facing the challenge of a continuous decline in both its number of subscribers and advertising revenues despite the efforts of traditional journalism to adapt to the needs of the online audience. Sometimes, revenues from the digital version, in the form of advertising and subscriptions, do not compensate for the losses of the print version [9]. The speed at which technology changes may be greater than ever in the history of journalism. Linden [30] points out that there is a strain on the editorial team, as adaptation is necessary for new tools and techniques and multimedia communication. In addition, journalists today face the automation of some facets of journalism in what some authors, such as Karlsen and Stavelin [28], designate as computational journalism. Therefore, this article proposes a cybermetric methodology based on linear regression models that optimizes the content publication strategy in digital media, which could help to improve the allocation of resources, the selection of content, and the measurement and optimization of the content strategy. This prioritization of information, already highlighted by Graefe [19] in his model of news generation, is raised in this study with a new prism: the design of a methodology that combines web analytics and social media trend data with machine learning models to assist in the editorial process. 2 Literature review 2.1 Digital journalism Adaptation to the online format has also enabled the monitoring of reader behavior and consumption. Over time, online newspaper websites have adopted digital marketing practices such as SEO (Search Engine Optimization), optimizing their content and platforms for search engines and using tools such as Google Analytics to measure traffic on their website [18]. This “measurabl

FAQs

What problem does the study address in digital journalism?

It addresses the need to better measure and optimize the success of online news content under changing digital conditions and limited editorial resources.

What methodology does the study propose to predict success?

It proposes a cybermetric workflow that selects indicators and tools, collects data, applies multiple linear regressions, and validates prediction equations to choose the most accurate model.

How is the study validated using the TV-series news use case?

The research uses an online media outlet and compares prediction equations by segmenting data into three sets: all articles, TV series articles, and trailer articles, then selecting the most accurate validated equation.

Aug 4, 2026

Upload by Valentina