WEKA - Experimenter

1
WEKA - Experimenter

• (sumber WEKA Explorer user Guide for Version
  3-5-5)

2
Applications - Experimenter
3
Experimenter - Setup

• The setup of experiments is divided into two
  parts
• Standard how to setup experiments in general
• Remote how to distribute experiments over
  several machines to speed up the execution time.
• New experiment click New
• Simple
• Advanced
• Results destination
• ARFF file
• CSV file
• JDBC database

4
Experimenter - Setup

• Experiment type
• Cross-validation (default) with the given number
  of folds
• Train/Test Percentage Split (data
  randomized)splits a dataset according to the
  percentage into a train and a test file, after
  the data has been randomized and stratified
• Train/Test Percentage Split (order preserved)
• Iteration control
• Number of repetitions default is 10.
• Data sets first/Algorithms first

5
Experimenter - Setup

• Example
• Try an experiment performs 10 runs of 10-fold
  stratified cross-validation on the Iris dataset
  using the ZeroR and J48 scheme, and save the
  result in exp1.arff file.

6
Experimenter - Run

• To run the current experiment, click the Run tab
  at the top of the Experiment Environment window.

7
Experimenter - Analyse

• After the experiment setup is complete, run the
  experiment. Then, to analyse the results, select
  the Analyse tab at the top of the Experiment
  Environment window.
• Results can also be loaded from an earlier
  experiment file by clicking File and loading the
  appropriate .arff results file.
• Click Perform Test to generate an output file.

8
Test Output