My latest commercial project: SenseArray
Monday, April 21, 2008 at 12:43PM I've just been putting the finishing touches to the website for my latest project, its called SenseArray. Its a stand-alone software application that companies can license, which allows them to easily integrate personalized recommendations into their service or website. I've been working with several companies over the past few weeks and months that are deploying the SenseArray technology, and I'm now ready to cast a wider net.
The basic idea behind SenseArray is simple: the success of almost any website or service depends on your ability to figure out what your users want, and then give it to them as expeditiously as possible. This can be the difference between success and failure, or between some success, and huge success. The problem is that its hard to figure out what your users want, and often they all don't want the same thing.
I'm not the first to try to tackle this problem, indeed, its not the first, nor the second time I've tackled this problem, as I developed two collaborative filters for Revver, and one for Thoof.
But building collaborative filters isn't easy, it takes time, and a lot of trial and error. I've done the hard work, and successfully navigated the minefield. The end result is that SenseArray scores more than 5% better than Netflix' own algorithm with the Netflix Prize dataset.
Furthermore, all collaborative filters suffer from a common problem, they all suck when it comes to new users, because they don't know anything about those users yet. SenseArray doesn't suffer from this problem, because it can utilize all kinds of data about users, much of which is available right from the very first moment they visit a website (such as their IP address, their operating system, and the referring website).
Anyway, take a look at the new SenseArray website, let me know what you think, and tell your friends about it if you think it might be useful to them!
Reader Comments (8)
Do you currently, or do you plan to, extend SenseArray recommendations with APML data?
Chad, it would be relatively straight-forward for APML data to be provided as input to SenseArray using the tagging mechanism described in the SA documentation.
Unfortunately it would not be easy to produce APML output from SenseArray because the way that SA internally represents a user's interests is very different from the approach employed by APML.
If there was significant customer demand for this we may be able to find a way to do it, but we have not had any requests for this from customers to-date.
[...] But in the last couple of days the site has been redirected to Reddit, and Clarke officially moved on to another project in [...]
Hi, Ian. I was a fan of Thoof and sorry to see it go but it seems like you're finding a new use for your technology. I run a website called The Daily Tube, which is a human-powered search service that tries to find the best new videos posted. Please take a look and tell me what you think. Would love your expert opinion.
We're planning to expand and I wondered whether you think your recommendation app would work for videos on our site?
Thanks,
Michael (michael@thedailytube.com)
Thanks so much for your professional thought about this post. But to find the masters’ essay writing services people must know some stuff about pay-essay.
I greatly appreciate all the info I've read here. I will spread the word about your blog to other people. Cheers.
Essay Service
Did you use an assistance of a custom sociology essays service for your hot outcome? I think that you have got great college essay creating technique. Thank you very much for sharing that!
I greatly appreciate all the information I have read here. I will spread the word about your blog with others. Cheers.