recommendations

Q&A community Fluther gets personal(ized)

Fluther, my favorite Q&A site has launched a new feature Wednesday called "Your Fluther." It lets you follow other people's activity on the site in one centralized, easy to parse feed. It's a companion to the built-in recommendation engine "just for you" that will feed you with questions based on topics listed in your profile and tracked site usage. More importantly, it's an easy way to create a private group of users who you'd rather keep an eye on than the growing public feed.

One thing I'd like to … Read more

Why Digg's new recommendation engine is a step backward

Announced Monday and now live for all registered users, Digg's new recommendation engine adds a new layer of social context to the upcoming section that lists stories dugg by other users and how much their reading habits match up with yours.

Like I said earlier this week, it's a two-fold effort: one to give the social-networking element of the site some love by giving users more exposure to like-minded people, and another to make sure the site's massive influx of submitted stories gets a little more attention.

The problem is, the new system does little to solve … Read more

Digg's recommendation engine slated for release this week

The social news/video/photo site, Digg, will be rolling out their long anticipated recommendation engine to users this week. The rollout will be starting with a random sampling of users and eventually moving towards the entire user base at the end of the week. Digg's recommendation engine will try to engage more users in the upcoming section of the site by recommending stories from similar users.

Digg's upcoming section, where stories wait for enough diggs in order to get promoted, has been very intimidating for users to jump head first into, given the volume of stories being submitted. It was fairly hard to sift through all of the crap in order to get to the real gems. Hopefully the introduction of a recommendation engine will make it a lot easier for users to find the quality submissions.

I believe that the addition of a recommendation to Digg is going to significantly help with the usability of the site and direct people towards the long tail of submissions. By adding a lot of new eyes to the upcoming section, Digg's homepage will certainly get a much fresher look than we have seen recently.

The other important factor to look at here is the apparent benefit that the recommendation engine will be giving to websites. Making content more discoverable on Digg means more traffic for the sites that the content has been submitted from. I suspect that more and more sites will begin to feel the Digg effect as a direct result of the launch of this feature.

The introduction of a recommendation engine should give Digg a nice jolt. I haven't seen the feature in action yet, but if it is implemented correctly, I see Digg becoming a much more valuable resource and frequented by a lot more people. Kevin Rose posted a couple of videos on the Digg Blog which you can see embedded below. The first is an overview by Kevin of the recommendation engine, complete with graphics and the second features Anton Kast, Digg's Lead Scientist, talking about the new engine. You can read his whitepaper on the subject here.… Read more

Digg's recommendation engine goes live

This week social-news site Digg is launching a new way for users to find new stories by adding a recommendation engine that will suggest things for users to read in the upcoming story section based on past site usage. It will take into account what other people similar to yourself are digging on the site and add a special note that shows how many stories you and that person share in common.

The move is two-fold: One part is to expand the social network that has been Digg since the launch of its expanded profile system back in September. The other is to give upcoming stories a little more attention. Stories submitted to Digg can reach the site's front page a number of ways, either by being linked up to highly trafficked external sites, or simply by natural selection in Digg's upcoming queue. The updated recommendation engine will simply be a new way for those stories to get more eyeballs on them given that the number of submitted stories averages around 16,000 a day, according to founder Kevin Rose.

The new feature is only being rolled out to a random number of registered Digg users and is expected to make its way to everyone else in the months to come. In the meantime, you can get some suggestions for stories you might like based on your previous "digging" habits using a third-party service called DiggSuggest. It doesn't use the same algorithm, or do it passively and onsite like Digg's does, but it comes with some pretty interesting results.

You can see the new feature in action in the explanatory video below. After the break, there's also a video with Digg's chief scientist Anton Kast explaining in detail how the new system works.

Digg Recommendation Engine from Kevin Rose on Vimeo.

Baynote ramps up offering with Collective Intelligence Platform

Baynote specializes in using traffic and user data in order to tailor recommendations and generate more sales. Launched in June 2006, Baynote employs this "wisdom of the crowds" approach and provides customized reporting to large companies. Today, Baynote is launching their Collective Intelligence Platform. On top of their prior capabilities, Baynote's web software now allows for automatic email customization, featuring product recommendations, based on collected data. In addition, new features for mobile, search, and customized statistics are also included in this release.

Personalized recommendations are something that has been wildly successful for sites like Amazon and Netflix … Read more

Turn Delicious bookmarks into recommendations with InSuggest

InSuggest, the recommendation engine I took a look at back in February, has just launched a service that scans your bookmarks to give you links for related reading. The new tool is a derivative of the Web site analyzer which would take any link you dropped in and give you recommendations of similar sites. In that case, users could add up to three sites to get more narrowed results, whereas this tool is focused simply on where you've been.

For now, it's limited to Delicious users, but if you've got an account there you can plug-in your … Read more

Richrelevance nabs $4.2 million for shopping-referral tech

If Amazon would have spun off its recommendation engine to Web publishers and retailers years ago, the company might have pre-empted the spate of new up-and-comers with similar technology.

But then again, if that were the case, the former chief of Amazon's data-mining and personalization group David Selinger wouldn't have seen a hole in the market and started Richrelevance. The San Francisco-based company, which makes software that delivers related product links to prospective shoppers, is in a hotly contested niche with start-ups Aggregate Knowledge and Loomia.

On Tuesday, Richrelevance was buoyed with a fresh investment of $4.2 … Read more

DiggSuggest tells you what you should read on Digg

DiggSuggest is a new service created by 20-year-old Digg user who has used the service's API data feed to provide a feature the popular social news site currently doesn't: suggested readings.

All you have to do is plug in your Digg username (or someone else's) and it will scour your past dugg stories to serve up suggestions for stories both popular and upcoming. The more you digg the better the results should be. Not only does it show stories, but it also groups together keywords related to the things you like.

That's about all it does … Read more

Songkick helps you discover new bands and their upcoming shows

If you're a person who loves live music but hates having to keep up on when your favorite bands are coming into town, there's a great new service for you. It's called Songkick, and it's been designed to help you stay on top of upcoming concert dates, as well as discover new music from your existing tastes. It's making a complicated process wonderfully simple, and I expect it to be the next big thing in live music in the same way that Last.fm and Pandora were with prerecorded music tracks.

To figure out what … Read more

MusicIP plug-in deejays your iTunes

Late Monday afternoon, music recommendation engine MusicIP released a beta of its MyDJ iTunes plug-in for Windows (available as a free download from the company site). I've spent the last few hours playing with the plug-in, and thus far I'm impressed. Like the company's standalone program, MusicIP Mixer 1.6, the MyDJ iTunes plug-in scans and analyzes your music library to ferret out similar songs and artists. But the standalone software requires you to organize music in MusicIP, then export the playlist to your media player software. The plug-in lets you automatically generate playlists without leaving iTunes. … Read more