Note that some classes exist (even if not in Spring, at least in base Java) http://imgur.com/sNk4mE2
The key thing to keep in mind with generative models based on real data is that just because results were based on random generation doesn't mean they can't match something real.
Ugh I hate it when people make this claim or link this article. If you follow one more article link deep, they say implemented some of the earlier code that came out of it, code that provided the real business value consistently.
Do you know what caused a team to win?
They Realized (1) people rate movies higher on specific days ( why would you want to implement that into your estimator? And (2) they realized that all the movies that Netflix was asking about we're movies the user was willing to rate, and because users tended to rate movies they like, this biased the results, so basically, they didn't implement the hacking the test
Ehhh the public user base perhaps isn't really so important in foss fps games because these games are most often used as LAN games when not everyone owns the same collection of games. Before tf2, Warsow was pretty much the only good fps LAN parties could use legally.
Xonotic and its predecessor Nexuiz are also 100% free and open source (git.xonotic.org). Even the media like models and sounds are also released with source links, which considerably different from what Warsow is doing.
There are plenty of realistic applications for this! For example, controlling the reticle vs aim of a fps character, selecting units easily in an rts, selecting spells while looking around in an mmo, simulating multitouch, simulating button clicks to reduce cts symptoms ( so one hand draws, the other hand decides if you Re drawing and to what intensity by moving in and out of an intensity circle. )
To stick with good first step approaches, look at ngrams and nwords.
Basically, you need a reasonable feature to match similarity on. N-words are pretty easy to construct, a 2-gram would be every pair of words used in a document.
Tf-idf is a good metric with that kind of feature, because it handles well the bias of frequent words like "the"
This is a bit of an oversimplification of data mining to the point where I am not sure it is useful. Most interesting data exists in only a subset of a large feature set, where most items are irrelevant to the similarity metric. Take movies for example, if you tried to find similar movies using all features, key grip names and minor actors would unrealistically mess up your similarity score. This relates to the "curse of dimensionality".
Many data mining approaches first use a feature selection or feature extraction approach. That is, an approach which finds the relevant feature subsets, or discovers the underlying features of the data set.
Inverse Image search and the solution to the Netflix prize both used feature extraction approaches.
I've been working on a plugin using the new sidebar extensions api, and I'm pretty sure this update broke the api :( Planning to file a ticket in the chromium bug tracker soon.
(Edit for the HN readers: This comment and following comments were posted in response to an earlier frontpage design.)
I posted this on shacknews but to duplicate the discussion over here on hackernews, here's my feedback:
The frontpage splash for this is really hard to read. It's obtusely written. It focuses on features, not the narrative. It focuses on buzzwords, not what it does. I kind of get it from your long video, but I doubt anyone has the patience to learn about this. There are a lot of social tree-based todolist applications out there, so to win in this space you have to be really really really focused on UX over functionality.
Best of luck though, I think you have some interesting ideas!
With MindWallet, you can socially or privately organize your plans and goals. Simplify problems by breaking them up into subtasks and coordinate with friends to complete tasks on-the-fly. MindWallet enables you to manage your to-do list instead of having it manage you.
MindWallet:
* Displays your current tasks and lets you break them into subtasks
* Communicates with friends within your to-do list
"Get addicted to MindWallet in 10 steps:
blahblahblah"
Make this be a direct link that says:
(Get started with mind wallet.)
Then, this link opens a view to facebook connect.
When the user completes that, show them an automatically-created first task list.
For the first time creating a task list, show them guide indication arrows to important CRUD operations (e.g., "Create your first goal =>)
Think of this like a videogame tutorial, you have to show them by making them do it. Continue down your list, implementing each task as a tutorial step that you hold their hand through. As an anecdote, my grandmother recently couldn't figure out how to click a recipe link on my facebook news feed. I said "Cool recipe" and had the link posted. She couldn't figure out that blue text was a link. You have to really help nontechnical users.
Yup, less is more. Especially on an application main page. If you explain it somewhere else, just put a huge pretty button to click to find that guide, and don't give them directions anywhere else except the FAQ/Documentation page you have.
ask for less from Facebook. I didn't even try it based on your permission requests. I'd recommend asking for just the minimum of what you need and then if they need to access a feature that requires more ask for it then. Do a progressive rights request or have some video explaining the service.
Right now with the simple design and just text most people won't offer you their Facebook rights.
The key thing to keep in mind with generative models based on real data is that just because results were based on random generation doesn't mean they can't match something real.