Teen’s Website Predicts What News Stories You’re Interested In

Earlier this month, Leo Jiang, a 17-year-old from Toronto, released a news aggregation website called OmniFeed. The website not only fetches popular content from over 2000 news websites, but also intelligently predicts what news stories its users would be interested in. In less than 2 weeks, OmniFeed has garnered over 14,000 users and over half a million unique hits.

When asked why he started OmniFeed, Leo replied “When you go on a typical news site, most of the stuff there you just don’t care about. OmniFeed removes all the clutter and displays exactly what you want to see.”


Unlike many other news aggregators that simply use keywords in their user’s previously read articles to display relevant news, OmniFeed uses a complex algorithm involving social graphs and public data on the web. OmniFeed looks for trends on websites such as Facebook, Twitter, and Reddit to predict what its users might like.


“OmniFeed also sees what similar users have read and assumes that you will also want to read those articles,” said Leo.


OmniFeed is currently available only as a website, though Leo plans on making an Android app soon.

About OmniFeed


At its core, OmniFeed is a feed reader similar to Google Reader. Users can choose from thousands of built-in RSS feeds or add their own.


On top of that, OmniFeed has 2 additional features to place it above similar services. The “Recommended” section uses the prediction algorithm to display what the user will likely be interested in and the “Popular” section displays the most popular news stories of the day, even if they don’t match the user’s interests.


In addition, OmniFeed uses Readability’s content extraction algorithm to allow users to read news articles in-site, as well as a summarization algorithm similar to Summly’s to provide a 1-2 sentence summary of each article. Leo says these 2 functions “still need a bit of work.”


Website URL: