"how to write my JAVA application by learning the XML from RM"

gfyang

Hi, I am beginning to integrate RM into my own application.
First of all,  I want to implement the following by  java codes,

<operator name="Root" class="Process" expanded="yes">
    <operator name="ArffExampleSource" class="ArffExampleSource">
        <parameter key="data_file"	value="../data/iris.arff"/>
        <parameter key="label_attribute"	value="class"/>
    </operator>
    <operator name="SimpleValidation" class="SimpleValidation" expanded="yes">
        <operator name="NaiveBayes" class="NaiveBayes">
        </operator>
        <operator name="ApplierChain" class="OperatorChain" expanded="yes">
            <operator name="Test" class="ModelApplier">
                <list key="application_parameters">
                </list>
            </operator>
            <operator name="Performance" class="Performance">
            </operator>
        </operator>
    </operator>
</operator>

This is what I know:

RapidMiner.init();

		//read the data
		Operator inputData = OperatorService.createOperator(ArffExampleSource.class);
		inputData.setParameter("data_file", "C:/data/iris.arff");
		inputData.setParameter("label_attribute", "class");
		IOContainer container = inputData.apply(new IOContainer());

		//build the model
		OperatorChain simpleValidate = 
			(OperatorChain) OperatorService.createOperator("SimpleValidation");
		OperatorChain applierChain =
			(OperatorChain) OperatorService.createOperator("ApplierChain");

		Operator naiveBayes = OperatorService.createOperator("NaiveBayes");
		Operator test = OperatorService.createOperator("ModelApplier");
		Operator performance = OperatorService.createOperator("Performance");

		applierChain.addOperator(test);
		applierChain.addOperator(performance);
		simpleValidate.addOperator(naiveBayes);
		simpleValidate.addOperator(applierChain);

Of course, it does not work. How could I really make the application run? That is, inputting data, validating naive bayes, and then outputting the accuracy?

Thank you very much.

steffen

Hello gfyang

There is a file called "rapidminer-<version>-tutorial.pdf" which you download from the rapidminer page at sourceforge (http://sourceforge.net/projects/yale/files/). In this you can find a chapter "Integrating RapidMiner into your application" where everything is explained.

If you have further questions, do not hesitate to ask.

regards,

Steffen

gfyang

Hi, Steffen,

Thanks a lot for the reply. I have found the tutorial, and it is very helpful.

I build a complete process. However, I do not know how to get the value of the testing result (not from the console).
For example,

	RapidMiner.init();

	Process process = new Process();

	//read the data
	Operator inputData = OperatorService.createOperator(ArffExampleSource.class);
	inputData.setParameter("data_file", "C:/data/iris.arff");
	inputData.setParameter("label_attribute", "class");

	//build the model
	OperatorChain simpleValidate = 
		(OperatorChain) OperatorService.createOperator("SimpleValidation");
	Operator naiveBayes = OperatorService.createOperator("NaiveBayes");
	OperatorChain applierChain =
		(OperatorChain) OperatorService.createOperator("OperatorChain");
	Operator test = OperatorService.createOperator("ModelApplier");
	Operator performance = OperatorService.createOperator("Performance");
	applierChain.addOperator(test);
	applierChain.addOperator(performance);
	simpleValidate.addOperator(naiveBayes);
	simpleValidate.addOperator(applierChain);

	process.getRootOperator().addOperator(inputData);
	process.getRootOperator().addOperator(simpleValidate);

	// how to get the accuracy?
	// double accuracy = ???

	return process;

The accuracy is stored in PerformanceVector, but how to println this double value?

Sincerely yours,
gfyang

steffen

Hello gfyang

process.run().get(PerformanceVector.class).getCriterion("accuracy").getAverage();

I you want to use the accuracy within the process, you have to implement your own operator (for this task there is also a chapter in the tutorial) and retrieve the performancevector from this operator's input.

regards,

Steffen

gfyang

Thank you very much