With the rapid spread of information enabled by social and digital media, journalists are challenged to deliver high-quality, nuanced journalism. TRENDIFY aims to optimize newsroom workflows and news quality by developing a suite of AI-driven media monitoring software tools, enabling journalists to mine and synthesize relevant information from digital data streams, including articles and comments on (news) websites, blogs and social media feeds.
The quality vs. quantity conundrum in journalism
News content audiences are highly concerned with information quality and value trustworthy, in-depth journalism. Focusing on quality rather than quantity offers market position consolidation opportunities to news brands and publishers.
But to do so, they must overcome the societal and technological challenges posed by the rapid, heterogeneous data flow of social and digital media. Concretely, journalists and editors face the difficult task of deciding which online trending content, trends and opinions they want to write about. Retrieving and researching this information is currently an enormous time and effort sink for journalists, as they often spend hours on the collection and analysis of large volumes of information and on linking it to underlying contexts.
Order from chaos, insights from data
A software-based solution to these challenges would meet journalists’ and editors’ need for information by streamlining the media monitoring process and enabling them to link trending content with broader trends and vice-versa. This would facilitate their ability to offer nuanced, in-depth journalistic pieces that also take a wide range of opinions and information sources into account. Through co-creation and user testing with journalists, important domain knowledge and editorial expertise are incorporated in the resulting software.
A real-time, birds-eye view of the digital information stream
The objective and innovation goal of the TRENDIFY consortium is to develop a series of software tools that:
- enable journalists to pick out trending pieces of content, such as articles, videos and tweets, with high reader engagement;
- facilitate the identification of underlying time-dependent themes that connect multiple trending content pieces;
- help journalists understand the alignment of opinions on these content pieces and trends.
Paving the way for high-quality, nuanced, data-driven journalism
The TRENDIFY consortium is composed of experts in the media industry, online user behavior, trend detection and opinion mining. It will combine cutting-edge machine-learning algorithms and technologies for textual analysis, with new methods to assess and foster content and opinion diversity. The resulting solution will empower journalists to create data-driven content that meets the informational needs of a highly demanding readership.
“TRENDIFY will optimize newsroom workflows and news quality by developing a suite of AI-driven media monitoring software tools, enabling journalists to mine and synthesize highly relevant information from digital data streams.”
AI-driven trend detection for the distribution of diverse, independent and relevant news content for the readers.
Trendify is an imec.icon research project funded by imec and Agentschap innoveren & ondernemen.
It started on 01.05.2020 and is set to run until 30.04.2022.Download as pdf
- Roularta Media Group
- imec - SMIT - VUB
- VUB - Artificial Intelligence Lab
- Project lead: Nick Dutry
- Research lead: Olivier Braet
- Proposal Manager: Tom Willaert
- Innovation manager: Steven Van Assche