5
$\begingroup$

I'm looking for implementations of online learners. I guess that is possible with AdaBoost. Where you train the model and then you modify it by adding later more training data. However you don't have to re-train the entire model. Are you aware of this online implementation of AdaBoost in java?

$\endgroup$
  • $\begingroup$ ML in Java... Why are you doing this to yourself? $\endgroup$ – Digio Feb 13 at 6:38
2
$\begingroup$

A good place to start for implementations of online learning is vowpal wabbit. It implements linear regression, logistic regression, several extensions of logistic regression for multi-class problems, neural networks, and matrix factorization. It also has several other nifty features, such as on-the-fly ngram computation and spell-checking.

I'm not aware of an online implementation of adaboost.

$\endgroup$
  • $\begingroup$ is there a R implementation of Vowpal Wabbit? $\endgroup$ – RockScience Apr 11 '14 at 4:21
  • $\begingroup$ Not really. It's a C++ program you have to compile into a binary and run from the command line. I've written some wrapper functions around it in R to pass data to VW and read back the responses, but it's klunky. If you have a mac, you can install VW via homebrew. $\endgroup$ – Zach Apr 11 '14 at 23:31
0
$\begingroup$

probably way too late, but check out Weka and Moa. Both are written in Java, and Moa has online implementations of the more popular algorithms, and can modify the algorithms that Weka has to be online

$\endgroup$
0
$\begingroup$

Massive Online Analysis (MOA) is a well-known software for implementing and testing online learning algorithms. This tool contain an online implementation of Adaboost algorithm called OzaBoost.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.