Google Makes Google Reader More Personal

Update: Google has also now personalized search improved search results based on search history. Google thinks the new changes will have a similar affect on users' reading experience.

In Google Reader, there is a new "Explore" section that appears under the "People You Follow" section. Within this section are the previously existing recommendations feature and a new one - "Popular items."

"We use algorithms to find top-rising images, videos and pages from anywhere (not just your subscriptions), collect them in the new Popular items section and order them by what we think you'll like best,"
"And to make it easier to find interesting feeds, we're moving recommendations into the new Explore section and giving it a new name — 'Recommended sources,'" she adds. "Like always, it uses your Reader Trends and Web History (if you're opted into Web History) to generate a list of feeds we think you might like."

Apart from the new Explore section, Google has also added personalized ranking to feeds. There is a new sort option called "magic," that re-orders items in the feed based on your personal usage and overall activity in Google Reader. This can be accessed by clicking "sort by magic," under the "feed settings," menu of a particular feed or folder.

"Unlike the old 'auto' ranking, this new ranking is personalized for you, and gets better with time as we learn what you like best — the more you 'like' and 'share' stuff, the better your magic sort will be," says Yang. "Give it a try on a high-volume feed folder or
It's going to be interesting using Google's new personalization features in Reader moving forward. It should be a good way to determine just how well Google really does know users. Unlike searches, which are prone to be much more random, Reader is a more intimate experience, where users frequently return to specific topics and sources. This does occur in search to some extent, but the very nature of searching means you're looking for something wherever you can find it in most cases (not counting just being lazy and searching for specific domains).