- Download MOA from here.
- Run MOA with
java -cp moa.jar -javaagent:sizeofag.jar moa.gui.GUI
- Click ’Configure’ to set up a task, when ready click to launch a task click ’Run’. Several tasks can be run concurrently. Click on different tasks in the list and control them using the buttons below. If textual output of a task is available it will be displayed in the middle of the GUI, and can be saved to disk.
Note that the command line text box displayed at the top of the window represents textual commands that can be used to run tasks on the command line. The text can be selected then copied onto the clipboard. In the bottom of the GUI there is a graphical display of the results. It is possible to compare the results of two different tasks: the current task is displayed in red, and the selected previously is in blue.
- As an example, compare two different classifiers, a Naive Bayes and a decision tree, using a prequential evaluation of one million instances generated by the default RandomTreeGenerator stream generator:
EvaluatePrequential -i 1000000 -f 10000 -l bayes.NaiveBayes
EvaluatePrequential -i 1000000 -f 10000 -l trees.HoeffdingTree
The prequential evaluation is different from the standard cross-validation technique. It is an online evaluation, that each time a new instance arrives, first it uses it to test and then to train.
- You can use streams from ARFF files. Datasets are available here.
EvaluatePrequential -s (ArffFileStream -f elec.arff)
Also, you can generate streams with concept drift, joining several different streams. For example, the SEA concepts dataset is generated joining four different streams with a SEA different concept each one:
EvaluatePrequential -s ( ConceptDriftStream -s (generators.SEAGenerator -f 1) -d (ConceptDriftStream -s (generators.SEAGenerator -f 2) -d (ConceptDriftStream -s (generators.SEAGenerator ) -d (generators.SEAGenerator -f 4) -w 50 -p 250000 ) -w 50 -p 250000 ) -w 50 -p 250000)
- You can continue learning how to use MOA with the tutorials (Tutorial 1 and Tutorial 2, Tutorial 3, Tutorial 4, and Tutorial 5), and with the Manuals ( MOA Manual and MOA Data Stream Mining). Enjoy MOA!