Weka Example, a simple classification of lines of text

Question

I want to classify really long lines of texts. Strings.

Here is the error I get:

Exception in thread "main" java.lang.IllegalStateException: No input instance format defined at weka.filters.unsupervised.attribute.StringToWordVector.input(StringToWordVector.java:681) at org.berlin.weka.test.TestWeka.main(TestWeka.java:61) 

Here is the code, but I keep getting exceptions, maybe this is not configured correctly.

 package org.berlin.weka.test; import weka.classifiers.Classifier; import weka.classifiers.functions.SMO; import weka.core.Attribute; import weka.core.FastVector; import weka.core.Instance; import weka.core.Instances; import weka.filters.Filter; import weka.filters.unsupervised.attribute.StringToWordVector; public class TestWeka { /* * java -cp %WEKA_HOME% weka.classifiers.meta.FilteredClassifier -t ReutersAcq-train.arff -T ReutersAcq-test.arff -W "weka.classifiers.functions.SMO -N 2" -F "weka.filters.unsupervised.attribute.StringToWordVector -S" */ public static void main(final String [] args) throws Exception { System.out.println("Running"); final StringToWordVector filter = new StringToWordVector(); final Classifier classifier = new SMO(); // Create numeric attributes. final String[] keywords = { "test1", "test2"}; FastVector attributes = new FastVector(keywords.length + 1); for (int i = 0 ; i < keywords.length; i++) { attributes.addElement(new Attribute(keywords[i])); } // Add class attribute. final FastVector classValues = new FastVector(2); classValues.addElement("miss"); classValues.addElement("hit"); attributes.addElement(new Attribute("Class", classValues)); final Instances data = new Instances("Data1", attributes, 100); data.setClassIndex(data.numAttributes() - 1); /////////// Instance instance = new Instance(10); instance.setDataset(data); // instance.setValue(testset.attribute(0),testset.attribute(0).addStringValue(obj.toString())); System.out.println("==>." + data.attribute(0)); instance.setValue(data.attribute(0), data.attribute(0).addStringValue("test1")); instance.setDataset(data); // Add class value to instance. instance.setClassValue(1.0); // Add instance to training data. data.add(instance); // Use filter. filter.input(instance); Instances filteredData = Filter.useFilter(data, filter); // Rebuild classifier. classifier.buildClassifier(filteredData); } } // End of the class // 

Answer 1

Can you try to set the input format in the following way:
//rest of your code
// Use filter
filter.input(instance);
filter.setInputFormat(data); //data instances that you are going to input to the filter
Instances filteredData = Filter.useFilter(data, filter);

For example, you can go to http://weka.wikispaces.com/Use+WEKA+in+your+Java+code and Delete the filter. (note that they set the input format of the filter before passing instances for filtering)