Run the data mining tool weka in MyEclipseįirst configure the java environment, set the path, install MyEclipse. ('J48 classification precision:'+(right/sum)) If(m_classifier.classifyInstance(instancesTest.instance(i))=instancesTest.instance(i).classValue())//If the predicted value and the answer value are equal (the classification column in the test corpus must provide the correct answer, the result Only meaningful) M_classifier.buildClassifier(instancesTrain) //trainingįor(int i = 0 i weka jar file has the correct path

tClassIndex(0) //Set the row number of the classification attribute (the first row is 0), instancesTest.numAttributes() can get the total number of attributesĭouble sum = instancesTest.numInstances(),//Number of test corpus instances Instances instancesTest = atf.getDataSet() //read in the test file InputFile = new File('D:\Program Files\Weka-3-6\data\') //Test corpus file Instances instancesTrain = atf.getDataSet() //read in the training file Double-click the Test file in the package Explorer, and then add four lines of code below the package wekaTest sentence:Īdd the following code to the Main function (under//TODO Auto-generated method stub):įile inputFile = new File('D:\Program Files\Weka-3-6\data\') //training corpus file

Right-click on the project name in the package Explorer and select the last item in the pop-up menu properties->select java Build Path on the left->Library page on the right->click Add External JARs.->Browse the directory where weka is located, and change the weka. Add a reference to weka in the project:ġ. Create a program file: Right->New->class on the newly created package, select the public static void main(String args) checkbox, and click Finish.Ģ. Enter the name in the Name text box, here is assumed to be Test. Create package: Find the newly created project in the package Explorer, right-click on it ->New->package. Click the Finish button (at the bottom of the dialog).Ģ. Create a project: click file->new->java project in the menu, and start any name in the project name of the pop-up dialog box. To use weka programmatically, the steps are as follows:ġ.

Weka jar file has the correct path