Tinder for news articles, by this we mean the ease of quickly browsing 100's to 1000's of articles to better build curated portfolios. Coupled with machine learning similar to how Pandora picks music for you.
Ninder is designed for curators to create briefings for multiple customers efficiently by picking articles from a list of prioritized articles.
With Ninder we want to streamline the process of creating a briefing. Not just by prioritizing articles to the end user’s preferences, but by mixing in some machine learning that will learn to associate the growing trends in the articles.
Every customer has a list of summarized news articles tailored to their approval and rejection history.
When scanning through the article list, a curator can approve or like an article which is then added to the briefing list while adding the associated tags to the learning pool for that customer and profile types. Rejecting the article removes it from the current list, and downgrades any tags that might be in the personas.
Allowing the system to learn beyond simple matches and find relevant data for specific personas.
Next steps are to finish the machine learning and polish the UI.