Recommendations & content discovery
Latest research reports and white papers on recommendations & discovery
More R&D reports
Designing client-side recommendations
Paper 274: Exploring how to provide responsive, personalised, and privacy-preserving TV and radio recommendations through client-side processing and user feedback.
Mood-based classification of TV programmes
Paper 232: How viewers describe programme mood, and how machine learning can classify seriousness, humour and pace with high accuracy.
Connecting current affairs with archive content through the Semantic Web
Paper 260: How archive content can be automatically connected to live news coverage to provide historical context, background and supporting material for journalism.
Machine learning
The cold start problem with recommendations
BBC iPlayer has a constant stream of new users and content. How can we make recommendations when there isn't enough information about either?
