Not OP, but I recently have set up Tiny Tiny RSS and did something similar. I find that there are two categories: High bandwidth feeds (your NYTimes, Reuters, etc.), and low bandwidth feeds (blogs/projects I like to follow, the occasional youtube channels, etc.)

If you curate it (I have worlds news, US news, politics, tech, science, etc), I find it's actually pretty nice to be able to scan through the headlines and see what interests you. It seems to augment my Reddit/HN feeds pretty well. I find that I can scan through all the headlines in ~3-4 minutes.

I also find it is incredibly nice for the low bandwidth feeds. I don't feel like I need to check those everyday, when I get an update it shows a notification and I can go straight through it. As a result, I am subscribed to a lot more of those, as it very easy to look through.

Fever had something like that, unfortunately it's not available any more as an RSS provider. Not sure if there are any other solutions, maybe even self-hosted that aggregate content.

I'm currently using http://nuzzel.com/ to aggregate my Twitter, Facebook and Linkedin feed and it does a great job of filtering out the "important" stuff that people talk about. Not just the noise that comes and goes.

I'm really happy with it, so happy that I even activated the email digest every morning and push notifications. It's pretty neat if you don't want to be drowned in a stream of information but also want to know what's going on.

I've experimented with using clustering algorithms to rank stories from about a dozen different RSS feeds to get results based on media coverage. I have some basic output set up here, it could use a lot of improvements but it seems like a decent way to sort national/international news without needing social media data: https://confabulator.io/newsclustering/

I've wanted to do this as well. How does your site rank stories within a cluster? I had always imagined that if I were to do this, I would order them by centrality, where the distance metric is Euclidean distance between the word histograms.

I'm doing graph clustering, where stories are connected by shared terms, so I'm ranking based on the weight of the edges to a story inside the cluster. The terms are weighted by frequency over all the stories in the system with the intention of prioritizing articles contain a lot of specific terms that look like they're related to the story.

I do use it, I actually have over 400 feeds in my personal collection. I have to manage my settings so I don't keep anything longer than a few days. I don't think it's necessarily something most people would enjoy. I only read a small percentage of the articles.

It's not that overwhelming to me because I've been using RSS a long time and I know what to expect, but I could see how it's not everyone's cup of tea.

I have Reuters RSS feed and the amount of story duplication (3 stories about the same from different angles)and the sheer number of news is really a problem. Plus RSS contains only one sentence summary.

Maybe RSS + local ML would solve the problem?