Two year old wine social network Snooth announced today that it is now powering wine recommendations for the 25,000 editor tested recipes on Conde Nast's food site Epicurious. Snooth says this is just the first of a number of big sites that its custom algorithm will power recommendations on. That cassoulet? Snooth suggests you serve a Montevina Terra d'Oro Syrah 2002 ($15) with it. Nice.

Wine with food has got to be one of the most familiar kinds of recommendations offline, but the online recommendation technology industry is a fast growing one. The belief is that quality recommendations will serve as searches you never knew you wanted to perform - helping users navigate from one logical option to another, possibly making more purchases as a result and hopefully being better served by the websites they visit.

A food site with good wine recommendations sounds pretty tasty to me. Snooth says its recommendations are based on ingredients, cuisine and cooking method.

Here's how it works. First, the keywords are parsed out of a recipe, then they are run through an extensive food dictionary and a long decision tree is then followed. Is it a soup, is it a salad, what is the primary taste? Beef and nuts tastes mostly like beef; beef and liver tastes mostly like liver. How the ingredients are to be prepared is determined by their proximity to preparation words in the recipe. Recipes with expensive ingredients will see more expensive wine recommendations, inexpensive ingredients (lobster vs. shrimp, for example) will yield less expensive wine suggestions. Goodbye old one-liners about "if you're eating chicken!"

Nibbledish, Cookstr, Chow? All cool recipe sites but no wine recommendations, much less very sophisticated ones. It's easy to see how recommendations can provide a competitive advantage in a niche like this.

How to Use it

InSuggest for bookmarks is very simple. You enter your Del.ico.us username, it looks at your archive of bookmarks and then recommends other similar pages you might like to bookmark. You can filter by one or multiple tags you've used. Up to 20 tag filter options are provided but you can enter any tag you've used in your account.

You can run anyone's Del.icio.us username through InSuggest and get recommendations, not just your own.

The interface is very nice, it's one of the best uses of Ma.gnolia's Thumbshot.org that I've seen yet, and the whole thing feels fairly smooth. In fact, it almost feels too smooth. The recommendations come to you very, very quickly. No where on the site, or in response to our email inquiry so far, can we find an explanation of how it works.

Despite that, it does seem to work well. There are a limited number of ways to parse Del.ico.us data, though, and we wouldn't be surprised if there's just a touch of caching going on. It's hard to say, but the end result is good. InSuggest developer Dennis Gustafsson was elected "Engineering Hero" by The Swedish Association of Graduate Engineers last year, according to search blog Pandia. So someone's seen behind Gustafsson's work and liked it.

Continued below.

Our Recommendations

Beyond some clarity around process and perhaps basic instructions on use, there are a few other things we'd like to see from InSuggest. The first is a feed for future recommendations. The display is in Javascript so we haven't been able to scrape it yet. We'd also love to see a Greasemonkey script for displaying InSuggest recommendations on top of the Del.icio.us bookmarking popup, archive page and item pages.

Other features that would be nice would be the option to input a Ma.gnolia username instead of just del.icio.us, tooltips to display full item titles that are too long for the basic display and the ability to exclude particular domains from future recommendations. Some sort of user feedback to inform recommendations should be doable.

Finally, the biggest fish in the pond when it comes to Del.icio.us recommendations is user recommendations, not just item level ones. We'd like to see other users be recommended, ideally with those who tend to find items of interest earliest privileged on the list.

That may be too much to ask for, though. It's hard to say. Feeds and user recommendations are the kinds of gifts that keep on giving, though, and are far more compelling than one-off recommendations.

For now, though, we think InSuggest for bookmarks is worth checking out. You could very well discover some things there that are just the kind of thing you've been looking for.

• SoundFlavor
• MusicIP
• Owl Multimedia
• Elbo.ws
• The Hype Machine
• SeeqPod Music
• Upto11
• 9ViBe
• Jamendo
• Goombah
• iLike
• Last.fm
• ChartU
• Mog
• MyStrands
• Qloud
• U.[lik]
• herkesdinlesin

If you've used any of the above services, let us know in the comments. I'd like to check out a few of these and see how their recommendations stack up.

As is typical for Apple, the company is not exactly transparent when it comes to describing how the 'Genius' feature actually works. It looks as if Apple compares your music selection to that of other users and then builds its recommendations based on this. We assume that iTunes looks at data about play and skip counts, beats per minute (which is available for all songs in the iTunes store), ratings, and playlists.

Because these recommendations are at least partly based on the libraries of other iTunes users, iTunes periodically downloads updated recommendations. You can also force an update from the 'Store' menu.

One fact that surprised us was that Apple often returned playlists for songs that were clearly mislabeled, which has led us to speculate if Apple, during the first run of Genius, actually creates an acoustic fingerprint for every song.

According to Apple, all the uploaded information is anonymized.

Does it Work?

In our tests, the recommendations and playlists were often spot-on, but also a bit inconsistent. Sometimes we would get great recommendations based on songs from rather obscure bands, while we sometimes couldn't get any recommendations based on songs from more popular and contemporary artists. For classical music, the recommendation feature basically didn't work at all.

We also noticed that the recommendations tend to favor more popular mainstream artists, but that could easily be a function of the current user base.

Apple points out that the recommendation engine will get better over time, as more users start uploading their information. It would be nice, however, if Apple also gave users a chance to tweak settings for themselves or at least gave us more information about how these recommendations are calculated.

One minor annoyance when using the recommendations is that if you decide to build a Genius playlist based on a song that is already playing, iTunes starts the song over after creating the new playlist.

What about Last.fm and Pandora?

As Last.fm co-founder Marting Stiksel pointed out in an interview with Wired's Eliot Van Buskirk, the 'Genius' feature basically validates what other music recommendation services have been doing for a long time.

It's also important to point out that a lot of other music recommendation services have strong, built-in social networking functions. Apple, even though it now has information about the listening habits of a large chunk of its users, does nothing to connect these users. One neat function, for example, would be for iTunes to show playlists from other users that have a certain songs in it. For now, though, it doesn't seem as if Apple is interested in adding these social aspects to iTunes anytime soon.

Rediscovering Music

For now, when it works, Apple's recommendations are actually a very nice way of rediscovering a lot of music that had long been sitting in our jukebox but never saw the day of light. We also assume that the shopping recommendations in the sidebar will drive more traffic to Apple's music store, especially once the recommendations get a bit better and users get comfortable with trusting Apple's recommendations.

The highlight of the new version is recommendations. The new Diigo offers a number of social networking type features that in-and-of themselves aren't worth a lot to me, but if they can do some number crunching and recommend people and content that I may want to subscribe to - that's gold.

What's the biggest crime committed by Del.icio.us? It's not leveraging the huge amount of data the service holds for some recommendations. Why on earth, in this data-centric era, isn't every social bookmarking service making bookmarking social and smart? If Yahoo! held an Amazon-style contest for recommendation algorithms that could be run against Del.icio.us, they could set up a Yahoo! News style page that was personalized like nobody's business. We'd all come back daily to read Del.icio.us, they could run ads up the wazoo and everyone would be beside themselves with happiness.

Instead we'll have to look to a pre-acquisition startup with neither network effect nor scaling problems. Diigo has potential to change the social bookmarking game just because they are offering recommendations. The recommendations aren't even very good yet because there's very few people using the service and the algorithm appears quite simple. I imported several hundred bookmarks from Ma.gnolia and perhaps Diigo will think deeper thoughts about my history after a few hours. I'm not so sure, though. It's still worth a look because it has so much potential.

You might also like the annotation features, though in all likelihood they will prove more trouble than they are worth unless you're an academic. You can associate an OpenID account with your Diigo account now, too. That's good.

Trust

Checking out Diigo could be pretty pain-free. The service does a good job of importing your bookmarks from elsewhere and allows you to publish simultaneously to your account at Del.ico.us, Ma.gnolia or Simpy. If, that is, you are willing to trust the Diigo people with the password to your usual social bookmarking account. Doesn't Ma.gnolia at least have oAuth support so I don't have to do that? Discussion about user authentication protocols as part of data portability seem common enough by now that it's outright offensive to be asked for your password to another web app. If you can deal with that, then there's no reason not to give Diigo a try.

Check out Diigo for yourself, it could be just what you're looking for. It's getting closer to something I can imagine using regularly and really appreciating - but it's not there yet. I'll keep an eye on the recommendations feature because if that ends up working out well, it would be reason enough to switch to Diigo.

Recommendations

In a tweet, the Noovo team told us that the app uses "state-of-the art deep graph mining and text extraction to recommend relevant content to users." A recent blog post explained further that Noovo uses "advanced algorithms to filter out the noise for you and recommends you the most relevant stories based on who your influencers are".

Further, Noovo recently deployed integration with dbpedia, the structured data version of Wikipedia. Noovo stated in its blog that this enables item-based recommendations, in other words pulling out topics from your user behaviour and that of your social network.

In many ways the recommendations part is like Digg's feature of the same name, in that it recommends interesting content from the site's community that you may like. And the more you use Noovo, the better the recommendations supposedly become.

Features Galore

Noovo is an interesting app and it sports a visually appealing interface. However, some of the main features are hard to find and then understand when first getting started. For example we had to hunt around to find out where the recommendations are (on the oddly named 'Cover' page, as it happens), and the hour glass icon is confusing at first glance (when you click it, it shows how the recommendations came about).

Adding content can also be cumbersome, unlike Tumblr where it is very simple and intuitive.

It's fair to say that these issues arise because Noovo has so many features - one could argue too many. But that also may end up its strength, because if you're looking for a central place to aggregate, share and discover cool content on the Web - Noovo could be a great choice for you. The community is small right now, but there is no shortage of colorful content to browse. Check out Noovo CTO Matej Pangerc's page, for example - you can see straight away that Noovo is very akin to Tumblr, Soup.io (my personal favorite) and other lifestreaming blog platforms.

We'll be keeping an eye on Noovo and testing it out some more. Let us know your thoughts in the comments.

CleverSet, now known as 'ATG Recommendations', is described on ATG's website as "a predictive recommendations service". It's just one part of a suite of e-commerce "optimization services". The company claims that it is differentiated from other recommendations services by its focus on commerce and the fact that it is a stable, profitable company. In the 2008 year the company did $164M revenue "with profitability". That figure includes many services and licensing revenues, of which product recommedations is just one. ATG was founded in 1991 and it did an IPO in 1999, so it does appear to be more experienced than competitors like richrelevance.

Predictive Recommendations

ATG's core product is an e-commerce suite, which it says is used to power hundreds of online store fronts. Its "e-Commerce Optimization Services", which includes recommendations, can be used on other e-commerce sites as well as those powered by ATG. The company told us that their products aim to increase the following 3 things: conversion rates; order value; and customer attention.

ATG calls its approach to recommendations a "blended approach", which aims to predict what the consumer wants to buy next. The recommendations come from a combination of user, site and product data. Elements include purchase history, the time of day, where the user clicked from, what browser they use, product catalog variables, historical shopping information, click-stream data, and more. Out of all this ATG provides what it calls "predictive recommendations".

How is ATG Different From richrelevance?

ATG's approach sounded very similar to that of richrelevance. So we asked ATG: what's different? Bruce D'Ambrosio, founder of CleverSet and ATG's Chief Scientist, responded that richrelevance is similar, but that it is "mostly a subset" of what ATG Recommendations does. He said that ATG is focused on bringing the merchandiser "into the conversation with the visitor", whereas richrelevance perhaps has less focus on merchandiser and more on the user.

D'Ambrosio further said that ATG models the visitor in both current and longer term sessions. The method it uses is called "Statistical Relational Learning" (SRL), whereby ATG integrates information about the actual user with "similar visitors". It also incorporates relational data about the product and "context of use".

Another key piece of data that ATG focuses on is "engagement", by which it means what fraction of users/buyers interact with recommendations. D'Ambrosio said that they see an enormous amount of this engagement activity before users buy products.

The Netflix Prize

As an interesting aside, we asked D'Ambrosio, as a very learned and experienced engineer in this field, what his thoughts were on the Netflix Prize. We mentioned the Napoleon Dynamite problem, whereby products like that are hard to recommend against.

D'Ambrosio replied that to win the Netflix Prize will require a re-definition of the problem, by "dramatically expanding the scope of the information". He said that most of the interesting information required to do Netflix recommendations well is actually off-site - e.g. product data that's not in the catalog on Netflix. He thinks that to win the Netflix Prize, contestants must gather more information about products like Napoleon Dynamite.

Conclusion: Big Player

As well as finding that it's a crowded market, another thing we've discovered in this series is that comparing recommendation engines directly to one another is nearly impossible. Firstly because each customer has different needs and so unless the customer has tried more than one product, they won't be able to accurately compare them. But secondly, each recommendation engine has a different approach and their algorithms are complicated - and well protected. Who knows what is really under the hood of Baynote, richrelevance or ATG.

However, ATG does have a good track record behind it; and we were particularly impressed by the knowledge of Bruce D'Ambrosio - surely one of the company's best assets in recommendations technologies. So if you're an e-commerce shop looking to implement recommendations, ATG seems like a relatively safe bet.

Does this mean Shazam is about to give Pandora and the like a run for their money?

The free Shazam application is best known for its nifty tune identification trick. Mobile users can hold their iPhones up next to a speaker or other source of music and the application "listens" to what's being played in order to identify the song and artist. It also lets you read track and album reviews, read artist biographies and tag songs to share with friends over Facebook and Twitter.

The new application, Shazam Encore, adds even more functionality including improved speed performance, trend lists that highlight what's popular among other Shazam users, a search function that taps into a database of 8 million+ songs, music recommendations and a "drive-and-tag" feature that lets the app recognize when it's in an in-car dock so it can identify what's playing on the radio while you're driving.

But How are Those Recommendations?

Out of all the new features, however, it's the music recommendations option which is the most interesting. Recommendations are the killer feature which can either make or break a mobile application these days. With services like Last.fm and Pandora already providing mobile users with playlists based on a user's likes or dislikes, Shazam needs to be able to do recommendations well - really well - in order to compete with these already popular applications.

In addition, the up-and-comer streaming music service from Spotify also partnered with The Echo Nest's music intelligence platform earlier this year to help improve on Spotify's playlist and music discovery functions. The end results of that partnership have been touted as being like the iTunes' "Genius" feature, only better. Although not yet available in the U.S., Spotify's mobile application is one of the most highly anticipated applications as it provides a new way to enjoy music - through playlist creations that can be listened to both online and off. It, too, will be heavy competition for any application entering into the music recommendations game, including, of course, Shazam.

So where does that leave Shazam Encore? At the moment, its recommendations offering provides you with a list of other songs you might like based on the one track you have pulled up. While this might help you discover new music, you aren't able to create a playlist based on those songs. Instead, Shazam's focus remains more on the sharing of music via tagging and posting to Twitter and Facebook.

As far as how good Shazam's recommendations are, we would need to do a lot more testing before giving a solid opinion - the app is just too new. In fact, it's so new that it wasn't even showing up in an iTunes Store search at the time of writing. The provided screenshot in the App Store doesn't look all that encouraging, though. (Really, a fan of indie band My Sad Captains wants to listen to Katy Perry singing about "kissing a girl?" I don't think so...)

But whether or not the recommendations are up to speed, it remains to be seen whether iPhone app shoppers will be willing to fork over the $4.99 US (£2.99/ €3.99) to have access to them, especially when there's no playlist option included.

Those interested in trying the new Encore application can find it now in the App Store by clicking here.

Loomia is similar to Sphere and another app we reviewed recently, Apture. These services all aim to serve up more clickable content options on media websites - which means more user engagement and time spent on site for publishers.

We spoke to Loomia CEO David Marks and asked him how Loomia compares to Sphere, which at first glance appears to have much in common with Loomia. Marks said that Sphere is trying to do "semantic classification", i.e. analyzing the content of an article and recommending further content based on the findings. However Loomia focuses more on the user and so it does behavioral type recommendations. This can result in a more diverse set of topics, because users typically have a range of content preferences. It depends on the article though, said Marks.

Loomia currently has 2 types of deployment:

• Content (e.g. WSJ)
• Video (e.g. Brightcove)

Marks told ReadWriteWeb that video advertising is currently selling well for big media publishers. Accordingly these publishers typically now want to drive users to their videos - and Loomia has a widget to do that.

Marks told us that a lot of their publishers are "dollar focused" this year, therefore recommendations have become more than just an interesting feature on a website - they can drive more advertising dollars. As an example, Marks told us that a media website's Finance section may sell out with ads, but its Politics section may not (fairly common in big media websites). But the Politics section tends to get bigger page views, so to address the imbalance Loomia's recommendations widgets can drive users from Politics to Finance.

We've been looking at how recommendations are being used in the retail sector a lot, and Loomia is a neat example of how the same technology can have real value for the media segment. Let us know in the comments what other recommendation technologies have caught your eye in publishing.

In our tests, Mufin often performed better than Apple's Genius feature, but for now, the plugin is only available for Windows.

Mufin creates a unique fingerprint for every song in your library when you start the plugin for the first time. You can choose if you want the recommendations to be based on the analysis of 30 second snippets (for fast analysis) or on the whole song (very slow, but highly accurate). Mufin's proprietary algorithms can then create playlists based on the similarities between the songs in your library. In creating these fingerprints, Mufin looks at over 40 characteristics, including tempo, instruments, and rhythm structure.

Apple's algorithms, on the other hand, are hidden in a black box, but seem to be based around the listening and purchasing habits of other users on iTunes.

Mufin's approach means that it will work for any song you may have imported into iTunes, no matter whether it is part of Apple's library or not. Mufin is also agnostic as to what language the songs are in.

Similar to Apple's Genius, the Mufin plugin will also make purchasing recommendations for similar songs that are not yet in your iTunes library and take you right to the iTunes store to listen to the preview or purchase them.

Verdict

Overall, we have come away very impressed with Mufin's recommendations. Judging from what we have seen so far, it may just replace the Genius feature as our preferred way of constructing automatic playlists.

That said, we are still not great fans of Mufin's core web service, which, unlike the plugin, is encumbered by licensing problems and which can only play 30 seconds of most songs (and often it can't play the songs at all). The plugin, however, is a clear winner in our eyes.

A couple of years ago, Alex Iskold outlined what he saw as the 4 main approaches to recommendations:

• Personalized recommendation - recommend things based on the individual's past behavior
• Social recommendation - recommend things based on the past behavior of similar users
• Item recommendation - recommend things based on the item itself
• A combination of the three approaches above

Amazon: King of Recommendations

In that post, Alex analyzed what Amazon.com - probably the canonical example of recommendations technology on the Web - used to power it's recommendations. Unsurprisingly, he found that Amazon used all 3 approaches (personalized, social and item). Amazon's system is very sophisticated, but at heart all of its recommendations "are based on individual behavior, plus either the item itself or behavior of other people on Amazon." What's more, the aim of it all is to get you to add more things to your shopping cart.

As Xavier identified, other newer Internet companies have tended to focus on specific methods of recommendation. For Pandora, it is a deeper analysis of the item (using its "gene" theory); Strands has taken a boatload of VC money to try and become the number 1 social recommendations provider on the planet; and Aggregate Knowledge is taking more of a behavioral approach.

Google: Focus on Personalized Recommendations

The most successful Internet company of this era has without a doubt been Google. It too has been using recommendation technologies to improve its core search product. There are two ways that Google does this:

1. Google customizes your search results "when possible" based on your location and/or recent search activity;
2. When you're signed in to your Google Account, you "may see even more relevant, useful results based on your web history."

So Google is using both your location and your personal search history to make its search results supposedly stronger. This is very much the 'personalized recommendation' approach - and indeed personalization has been a buzzword for Google in recent years. However, the two other types of recommendation are also present in Google's core search product:

1. Google's search algorithm PageRank is basically dependent on social recommendations - i.e. who links to a webpage;
2. Google also does item recommendations with its "Did you mean" feature.

There are surely other ways recommendations technologies are being deployed in Google search - not to mention the range of other products Google has. Google News, its start page iGoogle, and its ecommerce site Froogle all have recommendation features.

ReadWriteWeb Resources for Recommendation Technologies

Let us know what types of recommendation technologies other companies are using. We also invite you to explore using our custom ReadWriteWeb Resources:

• RWW Recommendation Industry Feed Favorites OPML file (save link)
• RWW Recommendation Industry Feeds - Best of Feed (copy and paste to your reader)
• Click to preview the above feeds before subscribing (pop-up window)
• RWW Recommendation Site Search (Visit and Bookmark)

LinkedIn Events offers not just event search, but recommendations based on the contents of your profile, sophisticated information about attendees and updates about the events in your LinkedIn update feed. Eight thousand events are already listed and event organizers can ad more.

The recommendations are key here. Recommendation is like the search you didn't even know you wanted to do - it's a great way to surface value from noise.

Unfortunately the events page is down at press time, but we look forward to its return.

We like LinkedIn alot here at ReadWriteWeb (it's one of the primary news sources for our new site about hiring activity) and we think Events is a great addition to the service. The events feature appears to be built on the OpenSocial platform, so there's a good chance that these features will be available in other settings beyond LinkedIn in the future.

MyBuys' Technology

The services approach means that MyBuys deploys a variety of algorithms and doesn't favor one approach - unlike, for example, ATG, which uses a method it calls "Statistical Relational Learning" (SRL). This is really the crux of the difference between MyBuys and the other companies we've profiled so far. The likes of richrelevance, ATG and Baynote all have a defining technology (usually patented) which for each is the foundation of its recommendations approach.

MyBuys has no specific algorithmic approach. Rather it appears to license technologies from companies such as Blue Martini, BroadVision, MarketLive and Microsoft. However MyBuys does still have a patent on the technology which brings all these disparate algorithms together - it calls it a "patented portfolio of algorithms".

MyBuys' recommendations are a javascript include for their clients' websites - i.e. the heavy lifting is done on MyBuys' servers. Their clients can see their stats in a MyBuys portal, and also summary stats are emailed to the clients.

Understanding Consumers

MyBuys has a team of people that focuses on site performance for its retail clients. This team - which works across all of MyBuys' client base - focuses on driving performance using a variety of tools and processes. They also do experiments for clients to find out what works best. Rosenblum noted that MyBuys is almost always paid on performance.

On its website, MyBuys says that it "creates deep consumer profiles based on both explicit information we collect from you and from shoppers when they sign up for alerts and implicit information we collect as shoppers interact with your site." Rosenblum claims that MyBuys "understands consumers at a deep level", whereas he said that its competitors don't necessarily do. He told us that the Baynote approach is "strange" because they don't focus on the individual, but rather the 'wisdom of crowds' (which he said is a 'lowest common denominator' approach). Further, Rosenblum claimed that many of MyBuys' competitors don't understand the product catalog, that they suffer from the "cold start" problem - i.e. with a new product there is no place to start, unless you know about consumer retail behavior.

Side note: we're sure that MyBuys' competitors would disagree with some of the above assertions, so we welcome feedback from them in the comments below. One thing we've found in this series is that each company in this space is very willing to talk down their competitors! A sign of a very competitive market.

Examples

On to examples of MyBuys' approach. One is World Market, a retailer of furniture and other goods from around the world. It has a 'May we recommend' section on its homepage, which Rosenblum told us is based on MyBuys' algorithms and what other people have done on the site. After the user hits the homepage, MyBuys tracks that user - they know where they came from, they pay attention to what the user clicks on next, and so on. On product pages, there are a variety of different recommendations on the right of the page under the heading 'More great finds'. The categories under this heading can differ (e.g. for some products there may be no 'featured' recommendation).

Another example is Golf Galaxy, a web retailer of golf gear. This has recommendations such as "Other great ideas" and "People also bought". It also serves up recommendations in the shopping cart: "You may also like".

MyBuys doesn't just do website recommendations, it uses email a lot. If they know the email address of the customer, they will send follow-up emails (e.g. if a user abandons the shopping cart). Rosenblum told us that this works very well, however he assured us that emails are 100% opt-in. He said that for every dollar MyBuys drives through the site, another dollar comes through the email channel.

Conclusion

So how effective are MyBuys' recommendations? According to the company, when recommendations engage consumers (i.e. a user clicks on a recommendation), they're 5 times more likely to convert than when there are no recommendations. Rosenblum told us that its clients see an increase of overall site revenue between 5-20%, which is a similar figure to that which other recommendations vendors have given us. The addition of email usually results in even higher conversions, the company claims.

As to how MyBuys compares to its competitors, as we've noted in previous reviews it's very difficult to make a judgment on that. However we're interested to note that a recommendations vendor can compete well in this market without having its own unique patented algorithm. MyBuys pushes the 'services' approach much more than the other vendors. We're sure that some of MyBuys' claims about the competition would be challenged, nevertheless it appears to be a successful business in the retail recommendations market.

Mufin.com, however, which launched today, uses a fully automated system that only takes the actual sounds of a song into consideration. In our tests, Mufin often returned good results, but the fact that it doesn't take genres or the quality of a song into account can make for a frustrating experience at times.

Eclectic Recommendations

There is something to be said for this completely algorithmic approach. While social recommendation services tend to return results that safely stay within the same genre as the original song, Mufin's choices are far more eclectic. The most closely related song to Rick Astley's seminal "Never Gonna Give You Up," for example, is a song in Bavarian by a German folk pop group. Mufin really seems to like German songs, by the way, as more than half of the songs it recommended as similar to Frank Zappa's "Muffin Man" were from a Sesame Street album in German. In Mufin's defense, though, all these songs were quite similar in style to the original song.

Social Recommendations vs. Algorithms

As MG Siegler points out, the real advantage of social recommendation engines is that they are very good at filtering out bad music (though one might also argue that this can lead to slightly boring recommendations at times). Mufin neither cares about the quality of the music nor its popularity or language; it only looks at characteristics like tempo, instruments, sound density, and harmony.

Limitations

Mufin has already built an extensive library of songs, but it only holds the right to a limited set of them, which means that you simply can't play a lot of the recommended songs. Those songs that can be played are limited to 30-second previews , which can make it hard to decide if a song is really good and similar enough to warrant buying it.

If Mufin had partnered with a service like Rhapsody, for example, users would at least have been able to stream a limited number of full songs every month. Every song on Mufin features a link to iTunes and the Amazon music store.

Overall, we think Mufin is an interesting experiment, but we are not convinced that its algorithms return better recommendations than the more social approach of its competitors or the classification system of the Music Genome Project that drives Pandora's recommendation engine.

Customers of Spanish bank BBVA will now be offered recommended products and services, individual and anonymized aggregate analytics and personalized goal setting and alert services, all based on their banking activities.

What's Interesting About This Deal?

Using the Strands Social Recommender technology, BBVA will be able to offer intelligent observations and suggestions for personal finance. A demo of the product shows, for example, that users of the system might be given interesting statistics about the financial activities of people in a particular demographic group, then asked whether they belong to that group. It's like having a private, personal, math-powered financial adviser available for your use on demand.

With interfaces for the iPhone, BlackBerry and Nokia phones - analytics and recommendations will also be available outside of the desktop web browser. This is the kind of heavyweight application to see coming from online recommendation services.

Privacy Concerns

How will bank customers feel about having their personal and financial details thrown into the collective pot for analysis of recommendations to other customers? We think it may take some getting used to, but that kind of information is undoubtedly being aggregated inside of banks already. The prospect of allowing users to benefit directly from their collective data is an appealing one.

Will the recommendations offered all point crudely toward buying more services from the bank? Given the huge war chest that Strands commands and the caliber of hires they've made over the last year, we hope that the company's banking recommendations and observations will prove truly useful and engaging for customers and not just for the bank's bottom line.

Only time will tell, but we've said for some time that in a world drowning in data - powerful recommendation technologies that help point towards personally meaningful information have huge potential. Financial services are the next frontier for these experimental technologies and we hope that Strands will disclose statistics in time demonstrating the impact their service had on the financial lives of users around Europe.

Disclosure: Strands is a RWW sponsor.