Today’s news consumers expect personalized content recommendations. This personalization is based on algorithms that predict what a user is most likely to click on, and often tends to generate a filter bubble: an endless loop of similar recommendations. To overcome this challenge, the NewsButler project will investigate and create a smart content stream recommender and editorial optimization engine that helps readers and editorial teams create a personalized, user-controlled news experience.
Fighting “fake news”
Tech giants have rich means to develop highly personalized digital tools. However, their broad approach to content curation can lead to widespread noise overload and disinformation. Publishers offering high-quality content must develop their own media consumption platforms that can compete by meeting the needs of demanding consumers in terms of both pricing and personalization.
New tool and pricing model
Harnessing the expertise of companies active in news product development, machine learning and human-computer interaction, the NewsButler consortium will further develop high-potential content personalization tools and combining them with human and machine intelligence. This holistic project will also explore new business and monetization models.
Where human intelligence meets AI
Outcomes will benefit three stakeholder groups: content readers, content editors and publishers. The NewsButler project will result in:
- relevant article and content stream recommendations;
- new news channels that incorporate chatbots;
- richer reader expectation insights and greater control over recommendation;
- an approach that maximizes conversion and retention;
- estimations of willingness to pay and conversion, and new monetization models.
Driving news diversity and quality
In addition to the NewsButler content recommendation engine, the project will also culminate in an editorial dashboard that analyzes and suggests content streams. It will also develop an intelligent newsbot that interfaces between the engine and the users. These outcomes drive diverse, high-quality news offerings, avoids filter bubbles and poses a sustainable alternative to the “fake news” phenomenon.
“NewsButler will create a smart content stream recommender and editorial optimization engine combining human and machine intelligence that helps readers and editorial teams create a personalized, user-controlled news experience.”
An intelligent, personalized news engine for consumer engagement in publishing.
NewsButler is an imec.icon research project funded by imec and Agentschap Innoveren & Ondernemen.
It started on 01.09.2018 and is set to run until 31.08.2020.Download as pdf
- ML6 - Skyhaus
- Roularta Media Group
- imec - ITEC - KU Leuven
- imec - SMIT - VUB
- Project Lead: Nick Dutry
- Research Lead: Celine Vens
- Proposal Manager: Nick Dutry
- Innovation Manager: Steven Van Assche