FeedHub

Jan tipped me off to FeedHub. This is a beta attempt at filtering lots of RSS feeds. I have to admit I’m not 100% what it tells me, but it appears to be doing some kind of text analytics on my feeds. It then personalize a structure around those feeds. The goal is to reduce RSS clutter and noise, so I can focus on the topics and subjects I want to (so they claim).

Here is what mSpoke, the creators of FeedHub, have to say about its inner workings in a blog post:

“Very simply, we learn about you based on the implicit usage of your personalized feed and any explicit gestures you choose to share with us. We use this information to distill a set of “memes” that describe your preferences. Each meme represents some characteristic of a post, like its topic, popularity in del.icio.us, or number of Diggs. Each meme also has a strength that indicates how predictive FeedHub expects it to be in choosing content you’ll like. As we learn about you, FeedHub automatically discovers new memes for you and strengthens or weakens memes appropriately.”

So you basically give FeedHub your feeds as a OPML file, it analyzes them for you, and then builds a profile of your interests that you can manage and customize.The basic building block of all of this is what they are calling a meme, or an extracted category.

I’m quite confused about the overall experience and how this really helps me make sense of the feeds I currently subscribe to. If anyone has more experience with it, I’d like to hear about it.