If you ask me, I’d say personalization is one of the next big steps towards a more semantic web. Everything we “like” on sites like Facebook or GetGlue gives others information about the things we’re interested in. If you get enough of that kind of data, as well as similar data from the people we’re connected to, you can effectively judge a person’s tastes and interests. Don’t believe me? How good is Netflix at picking out movies you’ll like?
Hunch [No Longer Available] is one such site that’s able to do this. Hunch personalizes the internet by getting to know you and then making smart recommendations about what you might like. In this article, I’ll show you how Hunch works, and why it offers the type of community that you just may want to be a part of.
What Is Hunch?
We last covered Hunch back in July of 2009, and originally in our directory back before we even put dates on posts in there. Back then we were calling Hunch a “decision making tool”. While that could still be considered a somewhat accurate way to describe Hunch, I’m going to call it what it is today: a personalized recommendations engine.
Here’s Hunch’s mission statement as depicted on their website:
Hunch’s ambitious mission is to build a ‘taste graph’ of the entire web, connecting every person on the web with their affinity for anything, from books to electronic gadgets to fashion or vacation spots. Hunch is at the forefront of combining algorithmic machine learning with user-curated content, with the goal of providing better recommendations for everyone.
Hunch provides personalized recommendations on tens of thousands of topics and is now partnering with other companies to power custom recommendations on 3rd-party sites and applications. It was started by “a bunch of MIT nerds” with backgrounds in computer science and math, who were exploring how machine learning could be used to provide smart, taste-driven recommendations.
How Does Hunch Work?
To get started using Hunch, head over to the homepage and sign in with your Facebook or Twitter account. Then you are asked a series of random (and I mean random) questions, which you can choose to skip or answer. After you answer each question, you will be able to see the percentage of people who answered the same as you. Answer as many as you like to better build your taste profile. It’s actually pretty fun, I must admit.
Hunch gets smarter/more accurate in two ways. First, since Hunch is powered by collective user knowledge, topics mature over time. Newly submitted topics often won’t be very smart at first, but as more and more people train and refine them, the topics will get much smarter.
Second, the more Hunch gets to know you, the more your recommendations will become customized. Every question you answer and topic you try helps this process.
When Hunch makes a recommendation, it’ll also show you why it proposed what it did. If you disagree with the reasoning, think it missed a crucial question or result, you can add all of that information yourself.
When others give one of your pros/cons a thumbs up, you get what are called Flecks. Flecks are like pats on the back, and people can “fleck you” for a question, result, or topic that you’ve contributed. You can give them to other people from their profile pages, or inline in a topic play. Written flecks must be approved by the person receiving the fleck before they are visible from their profile page.
Cred, Badges & Banjos
Sticking with the theme of getting/giving out props to people, you also can build Cred in the Hunch community. Cred stands for credibility, which is a summary of your Hunch contributions.
You also receive badges while using Hunch. Badges represent all the different ways you’ve contributed. Banjos (that’s right, Banjos) are one type of badge, which represent a numerical summary of your total contributions. Other badges represent the type of content you’ve contributed.
Also make sure you check out Hunch’s other stuff – like Twitter, iPhone apps and a Facebook game – over on their Goodies page [No Longer Available].
I think Hunch is a really interesting community. After answering just a few questions it recommended some of my favorite movies and TV shows. Once you use it for a little while and it learns more about you it can be a really useful tool for you.
What do you think of personalized recommendation engines? Are you going to check out Hunch?