Quickstart

This short How-To guides you from downloading the hyperSMURF weka plugin and load it into weka.

  1. Download the current stable 0.3 release from our GitHub project by clicking here.
  2. Download the latest 3.9 development version of Weka here and install it.
  3. Open Weka and navigate to the Package Manager (Tools -> Package manager).
  4. Install the hyperSMURF-0.3-weka.zip file (Unofficial File/URL -> Browse -> OK).
  5. Restart Weka. If you open the package explorer again you will see the hyperSMURF classifier under the Installed tab.
  6. Download example data quickstart_example.arff.gz
  7. Open the Weka Explorer. In the Preprocess tab click Open file... then navigate to the dowloaded quickstart_example.arff.gz and select it. Choose as filetype *.arff.gz and open the data.
  8. Now you should be able to switch to the Classify tab and choose the hyperSMURF classifier under weka -> classifiers -> trees. After that you can start the classification and the classifier output should display results without errors. The end of the output should look like:
Time taken to build model: 7.6 seconds

=== Stratified cross-validation ===
=== Summary ===

Correctly Classified Instances        8744               87.44   %
Incorrectly Classified Instances      1256               12.56   %
Kappa statistic                          0.7322
Mean absolute error                      0.2089
Root mean squared error                  0.2881
Relative absolute error                 49.7147 %
Root relative squared error             62.8514 %
Total Number of Instances            10000

=== Detailed Accuracy By Class ===

                                 TP Rate  FP Rate  Precision  Recall   F-Measure  MCC      ROC Area  PRC Area  Class
                                 0,823    0,006    0,997      0,823    0,902      0,758    0,993     0,997     c0
                                 0,994    0,177    0,707      0,994    0,826      0,758    0,993     0,985     c1
                                 Weighted Avg.    0,874    0,057    0,910      0,874    0,879      0,758    0,993     0,993

=== Confusion Matrix ===

        a    b   <-- classified as
        5759 1239 |    a = c0
        17   2985 |    b = c1