Java: Data Science Made Easy

Building the model

Before we can make any predictions, it is necessary that we train the model on a representative set of data. We will use the Weka class, MultilayerPerceptron, for training and eventually to make predictions. First, we declare strings for the training and testing of filenames and the corresponding FileReader instances for them. The instances are created and the last field is specified as the field to use for classification:

String trainingFileName = "dermatologyTrainingSet.arff"; String testingFileName = "dermatologyTestingSet.arff"; try (FileReader trainingReader = new FileReader(trainingFileName);         FileReader testingReader =              new FileReader(testingFileName)) {  Instances trainingInstances = new Instances(trainingReader); ...