You know how some posts go viral while others get completely ignored? This project figured out the hidden rules behind that. Researchers from 6 countries studied how information travels through social media, news sites, blogs, and email — basically reverse-engineering the mechanics of online attention. Think of it like mapping the invisible highways that rumors and news travel on, so you can either speed up important warnings or slow down harmful misinformation.
Companies in media, marketing, and cybersecurity struggle to understand why some online content goes viral while other content is completely ignored. Without predictive models, news agencies waste resources on stories that don't reach audiences, security teams can't ensure critical warnings spread fast enough, and marketers can't reliably predict campaign performance. The cost of misinformation spreading unchecked — from fake health claims to undetected cyber threats — continues to grow.
The project produced 22 deliverables including a metadata annotation and adapters tool for connecting multiple data sources, a complete data storage infrastructure with access capabilities, and training materials. The core output is a set of methods and models for predicting information spread, tracing information sources, and uncovering hidden communication channels across social media, news, blogs, and email.
If you are a news agency dealing with unpredictable content reach and declining audience engagement — this project developed data mining methods and metadata annotation tools that help you understand why certain stories spread virally while others are ignored. The Slovenian Press Agency was directly involved as an industry partner, validating these methods against real newsroom challenges across a 14-partner consortium.
If you are a cybersecurity firm struggling to track how warnings about cyber-attacks spread online — this project built data infrastructure and machine learning models specifically designed to trace information sources and uncover hidden information channels. With 22 deliverables including data storage infrastructure and metadata annotation tools, the research addresses how critical security alerts can be promoted more effectively.
If you are a marketing agency that needs to predict which campaigns will gain traction and which will fall flat — this project developed complex systems models for predicting information spread across different media and topics. The research combined social science, data mining, and complexity science from 6 countries to map how content dynamics actually work beneath the surface.
RENOIR was funded with EUR 1,273,500 under MSCA-RISE, which is a researcher exchange scheme — meaning most budget went to secondments and knowledge transfer, not product development. Licensing terms would need to be negotiated directly with the coordinator (Warsaw University of Technology) or the specific partner who developed the tool you need. Based on available project data, no commercial pricing model was established.
The project built dedicated data storage infrastructure (both early and final versions) and metadata annotation adapters for multiple data sources, suggesting the tools were designed to handle significant data volumes. However, the consortium was 86% academic (12 out of 14 partners are universities), so real-world stress testing at commercial scale is not confirmed in the available data.
IP ownership typically follows the Horizon 2020 grant agreement, where each partner owns the results they generate. With 14 partners across 6 countries including institutions in the US, Singapore, Australia, Russia, Poland, and Slovenia, IP is likely distributed. Contact the coordinator at Warsaw University of Technology to clarify licensing for specific outputs.
The project produced 22 deliverables including a Meta-Data Annotation and Adapters tool for Data Sources, training materials on data infrastructure, and both early and final versions of a Data Storage Infrastructure and Access system. These suggest functional research software rather than commercial-grade products.
The project ran from 2016 to 2019, which means the core research predates major shifts in social media dynamics (e.g., TikTok's rise, AI-generated content). The underlying methods for modeling information spread and finding hidden channels remain relevant, but the specific data and models may need updating for today's platforms.
The project's work on detecting misinformation sources and tracking hidden information channels is directly relevant to the EU Digital Services Act and similar regulations requiring platforms to address disinformation. Based on available project data, no specific compliance tools were built, but the methods could support regulatory compliance efforts.
The RENOIR consortium of 14 partners spans 6 countries including major research hubs in the US (Stanford, Rensselaer Polytechnic), Singapore (Nanyang Technological University), and Europe (Warsaw and Wroclaw Universities of Technology, Jozef Stefan Institute). However, with only 1 industry partner (the Slovenian Press Agency) and a 7% industry ratio, this is fundamentally an academic collaboration. The 2 SMEs in the consortium suggest limited commercial orientation. For a business looking to adopt these results, the strong academic pedigree means solid scientific grounding, but the path from research output to commercial product will require significant additional investment and development partnership.
Warsaw University of Technology (Politechnika Warszawska), Poland — contact through the university's technology transfer office
Want to explore how RENOIR's social media analytics methods could help your business track information spread or detect misinformation? SciTransfer can connect you with the research team and help assess fit for your use case.
