Predicting value: what analysis (Bayesian?) Can predict how much a Twitter user will rate a tweet?

Question

I'm thinking of adding a feature to the TalkingPuffin Twitter client, where after some training with the user, he can evaluate incoming tweets according to their predicted value. What solutions exist for a Java virtual machine (Scala or Java preferred) to perform this kind of action?

Answer 1

It’s not easy to do this well. Google would like to be able to do such things ("What links will cost the user"), like Netflix ("What movies will they rate") and many others. In fact, you should probably read the winning entry notes for the Netflix Prize .

Then you need to extract a bunch of functions, as @hmason says. And then you need an appropriate machine learning algorithm; you need either a function approximator (where you try to use your functions to predict values ​​between, say, 0 and 1, where 1 is the “best twist ever” and 0 is “omg who care”) or a classifier (where you use your functions to try to predict whether it is “good” or “bad” tweets).

, , "" ( ) - , Java.

, ; LIBSVM, (, ).

Answer 2

, , , y (x), , "x", , "" " ".

(Naive Bayes !), :

-, , , , - , , .