RSS readers can reduce information overload, but they also can become part of the problem as feeds add up. Startup mSpoke has an answer to RSS clutter with FeedHub, a new personalization tool that puts a content relevance knob at your finger tips.RSS readers (Bloglines, Google Reader, MyYahoo) work great with 10 or 20 information feeds, but there comes a point of diminishing returns as the number of feeds increases. About 30% of Web users make use of RSS feeds, according to mSpoke. These tend to be power users -- and the ones most in need of help managing all their news and blog feeds.

FeedHub is aimed at this crowd. After controlled alpha testing, it's now generally available in a beta version. To use it, go to the FeedHub site, and follow a two-step process: Upload your OPLM file, then subscribe with your preferred feed reader. FeedHub becomes the filter through which your RSS sources pass.

FeedHub applies machine learning to your RSS habits, applying greater weight to most-read topics and less to those you tend to ignore. As such, it's an automated tool. "We make your feed reader smarter," says Sean Ammirati, VP of business development and product management for Pittsburgh-based mSpoke. The technology behind it is mSpoke's mPower Adaptive Personalization Engine.

mSpoke has put control knobs on FeedHub so users can assign different levels of value to different topics. Key here are "memes," predefined topics and categories, such as the most popular Digg entries. So you've got machine learning (the mPower engine), human clustering (Digg, Del.icio.us), and user preferences all combined into one RSS filtering system.

One application that sounds promising to me: Accessing news and blogs through a FeedHub-enabled smartphone, a potentially fast and easy way to get your most significant news feeds while on the road.

mSpoke hopes to make money through advertising, but that requires volume, so first things first. "No. 1 now is to create a compelling experience for users," says Ammirati.