Problems programming my own tool

classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view

Problems programming my own tool


I want to write a tool which should process a dataset like the one specified
in the following flow.

Arff Loader --(dataSet)-->
Remove --(dataSet)-->
Discretize(US) --(dataSet)-->
ClassAssigner --(dataSet)-->
CrossValidationFoldMaker --(trainingSet/testSet)-->
NaiveBayes classifier --(batchClassifier)-->
PredictionAppender --(dataSet)-->
Arff Saver

My problems are these: I have generated the training and test data with the
CrossValidationFoldMaker, but now I dont know how to use them with the
NaiveBayes classifier. The method buildClassifier takes only one set of
instances as argument, the trainingSet. How do I use the testSet with it?
And the next question that arises is how do I generate a batchClassifier out
of the NaiveBayes classifier?
Im not very into programming, so I would appreciate if someone could show
short segements of code rather than plain explanations.

Thanks in advance


Wekalist mailing list
[hidden email]