Ask HN: Is there a recommender system for blog posts or articles?

14 points by samlhuillier ↗ HN
Something like Tiktok but just for aggregating links to blogs or articles

12 comments

[ 3.1 ms ] story [ 52.6 ms ] thread
I haven't found one either, but this is something I would very much enjoy.

I think one possible way to implement such a system would be to keep track of what people highlight and run some data analysis on that. Opt-in, of course, if you don't feel comfortable sharing that data.

I'm sure there's lots of great essays I've missed that are just waiting to be discovered again.

That hype was 10 years ago. Most of the companies doing variation of what you’re asking disappeared.

The three major remaining are google news, apple news and reddit, all manually curated.

I don’t think there’s a lot of opportunity, users simply want free content, you can’t really monetize through ads unless you have a massive user base (ie user generated content), and acquiring content is really hard.

If you want to give it a try yourself, I recommend an algorithm à la reddit/hn and some basic clustering to avoid duplicates. You can get scores from twitter/fb instead of relying on your own users. The hard part imo remains scraping content and parse the text… well, and monetize of course.

I think stumbleupon might be what you're looking for. It got moved to mix.com just btw