Furthermore, the interest graph has much more expressive power in the case of the mobile web. Many mobile interest indicators simply don’t exist on the web, for example, checkins, location data, near field communication, etc.

Let’s take a close look at what’s involved when building a complete and expressive interest graph platform, and how the resulting graph can effectively optimize ad/reward targeting.

Companies developing interest graph platforms have to overcome the following fundamental challenges.

1. Data Collection

There are two approaches to collecting interest data: explicit and implicit. Companies such as Hunch, Blippy and Pinterest have all attempted to address this by asking users to share interests explicitly, with the incentive that users will get more accurate recommendations in return.

However, this approach has some psychological hurdles. Why would I tell an engine that I like BMWs, only for it to suggest BMW deals to me soon after? And if I mention that I like sports cars, in general, would it suggest cars that are too pricey or too big for my taste? In order to get accurate recommendations, I should be more specific: I like sports cars, but mostly European ones, and only those that are under $60,000.

In other words, the amount of data that I have to volunteer in order to get a decent recommendation or deal is so high that I might as well just search for the product directly.

Amazon does a great job at implicit data collection. The site has all the elements of the equation: your purchase history, product search history and even product correlations (people who bought this also bought that).

Any app where users have to take the time to populate their own interests will invariably have inaccurate or artificial interest profiles.

Facebook encountered this issue with its interest hubs: Most people didn’t take the time to populate their profiles with accurate interests. Even if they did spend the time, they populated the subset of interests that they wanted to project publicly. Google+ Sparks faced the same issues.

Interests have to be inferred from normal app usage, where users have opted in. And the app’s normal usage has to provide strong interest indicators. General social networks such as Facebook and Google+ have the luxury of collecting vast data of various types. Specialized social networks such as Foursquare and Pinterest collect data that is heavily biased towards one signal (checkins in the former, and liking photos that belong to certain interest categories in the latter).

2. Noise Filtration

Every action online is considered a signal, and almost every signal in the digital world has its fair share of noise, though the noise levels and types vary greatly. For example, comments are extremely noisy (LOL, OMG, etc.), as are Likes/+1s when applied to photos or comments etc.

However, Likes/+1 of brand pages, for example, are very reliable interest indicators. Essentially, “noise” becomes valuable, depending on the amount of effort a person puts in. But the degree of correlation varies depending on the person’s behavioral profile.

If you take the time to upload a video of yourself skiing, that’s a strong signal. If you simply Like/+1 someone’s skiing photo, it may be that you like the person, or you like skiing, or you are simply trying to get the attention of the poster in order to start a conversation.

Repeated checkins at restaurants/bars are strong interest indicators as well. And clearly, reward/deal redemptions and purchases are very strong signals.

Machine learning algorithms are typically used to detect noise patterns and spammy comments, etc. Similarly, signal strengths have to be analyzed. To give simple examples, uploading sailboat photos every week is a different level of interest than liking a friend’s sailing photo once in a while. Similarly, checking in at the same sushi restaurant six times last month sends a very different signal than checking in once every two months.

However, here’s where the complexity arises. One checkin per month may actually be a strong indicator if the person travels frequently. In other words, the signal strengths calibration algorithm has to be customized according to the person’s behavioral patterns and lifestyle.

The beauty is that you don’t have to get it right the first time, since, like any neural network, the engines improve with usage over time.

3. Building the Interest Graph

Even after noise filtration engines have been well-trained and continuously re-calibrated to build an interest profile for a given user, constructing an interest graph for a set of users is still challenging. Essentially, the complexity is in aggregating all the signals to form a coherent and reasonably consistent profile.

4. Platform APIs

In my mind, this is the biggest challenge, and so far, no company has managed to deliver a solution. Clearly, a lot of companies build their own interest graphs for their own user bases.

However, being an interest graph platform means publishing APIs that any app/game can use to personalize ads or commerce to their users. This means that your APIs have to be at such fine granularity that other apps/games can integrate them seamlessly, with no detrimental impact on the user experience.

5. Distribution: Attracting the Apps to Use the APIs
As in any B2B sales cycle, the first few customers are the hardest to acquire. In this case, they are also the most important, since their impact on engine accuracy is very significant.

It’s very clear that interest graphs are at the core of the curated mobile web, and will be a key driver for mobile commerce for years to come. Many apps are already building relatively accurate interest graphs of their user bases. Ultimately, the companies that build scalable interest graph platforms with APIs that map to numerous apps and games will dominate the mobile commerce ecosystem.

VuHunt castles & cool places to compete for to earn the Emperor of San Francisco

VuHunt on iPhone

Congrats Emperor!
So you’ve joined VuHunt, conquered a few castles, and you’re now on your way to earning your first title. Prince/Princess of a cool city C, then onto the Duke and finally Emperor. We wanted to share a few tips and tricks in the game that aren’t yet in the tutorial..

1. Build it.
Build castles! it’s good karma, shows you are a great VuHunt citizen, and also earns you Vu$, and the honor of having your name as builder of a castle forever. Think about it this way, say you create an easy quiz castle, you get vu$2 to boot, then you can conquer that castle, and earn another vu$ 5, so that’s $7 for one simple contribution. Similarly, if you create a castle at say a Starbucks, make it a checkin challenge, and checkin right away, you’ll gain a comparable amount. Clearly, contributing castles improves the game for everyone, as the game world gets richer and more reflective of real-world places and captures the local trends and interests of the city’s residents.

2. Expand your empire! and dominate your country
Everyone likes a bigger powerbase. Ask any politician! It’s true in vuHunt as well! In fact, the # of castles you own isn’t only important for gaining the honor titles (prince/duke/emperor) and earning you a coveted spot on the leader-boards, but it’s crucial for growing your wealth. Each castle you own generates recurring revenue as long as you own it. That contribution is the sum of the daily net gain of the castle (revenue minus upkeep) in addition to the failed attack fees that you collect. This can be really significant income for your treasurey.

3. Protect your castles
One truly cool way to both protect a castle that you’ve struggled to conquer (either a difficult challenge or a steep travel cost) is to protect it with a guard battalion. This allows you to build 2 towers and hence two challenges for this castle. Now think about this as both a deterrant, and a source of additional revenue. Some players may just not want to go through the hassle, associated risk, and extra cost of attacking two towers. But also, it’s clearly highly likely that the revenue due to failed attacks will increase.

Guard battalions cost Vu$ to purchase so this techniuqe requires investment, but as you can see if can payoff handsome dividends.
Another mechanic to protect your castle, is to simply replace the challenge after you conquered it. Say it was an easy quiz qusetion, well you can purchase a challenge that’s a checkin or video upload and have the tower carry that challenge instead.
This only cost you the challenge purchase cost (typically 5 or 7 vu$)

4. Earn Real-life Rewards & get access to cool deals when you want them
So far, Vuhunt has integrated rewards from Kiip and daily deals from Groupon and Zozi. The kiip rewards are earned upon achievements such as building castles, gaining an Honor Title, etc. The deals are accessible through the Deals tab. At levels 2 and 3, which will be released for iPhone and Android very soon, you will be able to redeem VuDollars for gift cards. Also, a shortlist of VuHunt recommended deals will be available to you based on your interest and location upon demand.

5. Meetup and play against friends!
We are also very excited to announce that we have just started weekly meet-ups with our wonderful players. These meet-ups are informal brainstorm sessions to hear your ideas and feedback on the game and continue to enrich the experience with game elements that you find interesting. And we’ll play against each other real-time. The meet-ups are held in one city in California each week.
Please join the Vuhunt group here: http://www.meetup.com/VuHunt/events/46117612/

Happy conquering and best of luck!
Moataz Rashad

Drum roll please…. After much blood, sweat and tears, Vufind is ready to launch its amazing and eagerly awaited brainchild, VuHunt, the mobile game for both iPhone and Android! While applications and sites with deals, checkins, photo sharing, and augmented reality are blowing up, VuHunt is the triple threat that incorporates and puts its own spin on all these angles to create a compelling real-life game that has it all.

VuHunt brings out the serious competitor in you by allowing you to take ownership of locations, or castles, on either a virtual or real map of your city. Nothing brings out your inner competitor defeating others to feel the pride of ownership, or in this case gaining and maintaining your status as ‘King of your Castles’. Take on your friends, or maybe even strangers, and beat them to rule all of their Castles!

Through various quiz question relating to your location, photo challenges, puzzles, and check-in’s you can conquer each castle. By defeating your friends and taking ownership of their castles, you could even start your own empire! Challenges range from easy to difficult, so start off slow if you want and work your way up to total Castle domination! The harder the challenge the easier to keep your castle away from those that are trying to take it from you! But surely harder challenges cost you more Vu$ to purchase, so you’ll have to build up your treasure chest to make your empire invincible.

Part of the beauty of vuHunt is that it can be played anywhere at anytime! Lying by your pool and don’t feel like moving to check-in somewhere? There are various other challenges you can compete in, and you can kick your friends out of their castle without taking a step. Also, exploring new and exotic places is made easy, educational, and exciting. Travel the globe and learn about the Egyptian pyramids or set up a challenge for a Church in Munich, Germany, all while lounging poolside!

Now the final crucial and exciting aspect of vuHunt is the deals. Not only can you have bragging rights over your friends, but you can actually get daily deals and offers in areas and activities that actually interest and pertain to you! Maybe even invite your friends to make up for knocking them off their high horse and taking their castle…

Sign up for the Beta version of vuHunt and get a jump start on your friends! Watch @moatazr give a demo of an early version of vuHunt to @scobleizer here  http://www.youtube.com/watch?v=MJqYYnTrnhk.

Available now for Android and coming soon for iPhone!

Check it out, Compete, and Conquer!

And Developers: stay tuned!! we’ll offer awesome opportunities to build your own apps and plugins on the vuHunt platform soon.

Happy Hunting!

Samantha Nielsen

In the crazy tech world of checkin’s, tagging, and photo sharing, Vufind is the exciting new comer on the scene, where you can explore all of the above and more! The Vufind app for both iPhone and Android is the next step in connecting visually with both old and new friends.

No more manual tagging and spending hours organizing photos, with Vufind’s app you can either upload a picture/video from your phone or snap a new picture/video and have Vufind recognize and tag the objects in it. These tags can then be used to share your photos/videos with friends and be connected with new friends. The number of objects/tags the engine recognizes is continually increasing, and in fact the engine learns from its mistakes as users flag recognition errors. Building on the basic set of objects we have now such as car, bike, and watch, we plan to soon release finer-detail objects such as SUV, sports-car, touring bike, and Rolex watch etc.

For example, if one of your hobbies is kayaking you could have photos of your kayaking trips, which would be tagged with water and kayaks etc. People could then search your photos using those tags, learn about your interest in kayaking, and get connected with other people that like kayaking.

These tagged photos/videos can also be used to do visual checkin’s. While textual checkin’s can just be added noise on Facebook and Twitter, visual checkin’s with Vufind provide so much more information to friends. People checking in with Vufind aren’t just sharing the name of a location, they could share what it looks like, or what the view is, or even what they are eating.

Also, you don’t have to just checkin with a location, you could check in with an object. If you happen to come across a really cool or rare car you can take the photo, have it tagged, and show it off to all your friends! Visual checkins with tagged photos on Vufind are definitely more fun to share and are more interesting and interactive for others to see!

Last but not least, Vufind allows you to keep up with other friends and people around you as far as what’s popular and trending, so you can join in the trend! Vufind has a live feed so you can see what kind of photos everyone else is uploading, and contribute. If it’s summer time and everyone is uploading their best beach photos, you can show off a picture from your own beach getaway.

Vufind allows you to see what’s going on with all your friends in a fun and easy way, and keep up and contribute your own great photos, so your friends can see exactly where you are and what you’re up to.

Check it out, start checking-in visually, and share your life’s interesting moments!

Download the new Vufind 2.1.5 app and start uploading and sharing the smart way!

Happy tagging!

Vufind team

Recently, we also added cool imaging features such as make a collage, view a slide show, etc.

VuAlbum has just crossed its first 1000 monthly users, we look forward to the first 100,000 and beyond.

Currently, we face a tricky tradeoff. The tradeoff is simplifying the user experience and UI of vuAlbum, at the expense of performance and a slow response time.
Some friends and advisors argue that users don’t like to think too much, and by asking them to select object categories, we’re asking them to do too much thinking (and extra clicks)
if we let users click on “tag all” and don’t specify categories, the app would have to attempt to search and recognize every object the engine is aware of, which is time consuming — our engine has close to a couple of hundred object detectors at the moment, and growing every week. This has a few implications:

1) It’s costly for us to have the engine look for two hundred objects in our photos, when you maybe only interested in natural objects (tree, flower, etc), or scenes (beach, sunset, sky, cloud etc)
2) It will take longer to look for all these objects (we are running on tens of nodes on Amazon EC2, not hundreds, or thousands)
3) It’s likely that if you search for 200 object detectors rather than 40 or 50, that you will get a lot more false positives (tags that aren’t in the photo)
4) Most heavily tagged photos (such as on Flickr for example) have on average about 12 tags that are objects, the rest are descriptive or higher level tags, such as beautiful, peaceful, funny, or travel, sports, etc.

The thinking behind having the user guide the engine to what class/category of objects they care about was to the user’s benefit– to get the photos tagged quicker, and to not produce tags that are irrelevant to the user. We thought that if you are interested in sunsets, beaches, and trees, you would’d care if we found cars, bikes, and mobile phones in your photos. Hence, we asked you to select a category or two. But, if the extra clicks are painful, we will oblige and eliminate them.

Visit our vuAlbum fanpage on facebook, http://tinyurl.com/2e87y6k and share your thoughts on what you prefer. We are always listening! Thanks.

Thanks

Vufind team