Weka Manual

• I The Command-line
  • A command-line primer
    • Introduction
    • Basic concepts
      • Dataset
      • Classifier
      • weka.filters
      • weka.classifiers
    • Examples
    • Additional packages and the package manager
      • Package management
      • Running installed learning algorithms
• II The Graphical User Interface
  • Launching WEKA
  • Package Manager
    • Main window
    • Installing and removing packages
      • Unoffical packages
    • Using a http proxy
    • Using an alternative central package meta data repository
    • Package manager property file
  • Simple CLI
    • Commands
    • Invocation
    • Command redirection
    • Command completion
  • Explorer
    • The user interface
      • Section Tabs
      • Status Box
      • Log Button
      • WEKA Status Icon
      • Graphical output
    • Preprocessing
      • Loading Data
      • The Current Relation
      • Working With Attributes
      • Working With Filters
    • Classification
      • Selecting a Classifier
      • Test Options
      • The Class Attribute
      • Training a Classifier
      • The Classifier Output Text
      • The Result List
    • Clustering
      • Selecting a Clusterer
      • Cluster Modes
      • Ignoring Attributes
      • Working with Filters
      • Learning Clusters
    • Associating
      • Setting Up
      • Learning Associations
    • Selecting Attributes
      • Searching and Evaluating
      • Options
      • Performing Selection
    • Visualizing
      • The scatter plot matrix
      • Selecting an individual 2D scatter plot
      • Selecting Instances
  • Experimenter
    • Introduction
    • Standard Experiments
      • Simple
        • New experiment
        • Results destination
        • Experiment type
        • Datasets
        • Iteration control
        • Algorithms
        • Saving the setup
        • Running an Experiment
      • Advanced
        • Defining an Experiment
        • Running an Experiment
        • Changing the Experiment Parameters
        • Other Result Producers
    • Cluster Experiments
    • Remote Experiments
      • Preparation
      • Database Server Setup
      • Remote Engine Setup
      • Configuring the Experimenter
      • Multi-core support
      • Troubleshooting
    • Analysing Results
      • Setup
      • Saving the Results
      • Changing the Baseline Scheme
      • Statistical Significance
      • Summary Test
      • Ranking Test
  • KnowledgeFlow
    • Introduction
    • Features
    • Flow Steps
      • DataSources
      • DataSinks
      • DataGenerators
      • Filters
      • Classifiers
      • Clusterers
      • Attribute selection
      • Evaluation
      • Visualization
      • Flow
      • Tools
    • Examples
      • Cross-validated J48
      • Plotting multiple ROC curves
      • Processing data incrementally
    • Plugins
      • Flow components
      • Perspectives
  • Workbench
    • Introduction
  • ArffViewer
    • Menus
    • Editing
  • Bayesian Network Classifiers
    • Introduction
    • Local score based structure learning
      • Local score metrics
      • Search algorithms
    • Conditional independence test based structure learning
    • Global score metric based structure learning
    • Fixed structure 'learning'
    • Distribution learning
    • Running from the command line
    • Inspecting Bayesian networks
    • Bayes Network GUI
    • Bayesian nets in the experimenter
    • Adding your own Bayesian network learners
    • FAQ
    • Future development
• III Data
  • ARFF
    • Overview
    • Examples
      • The ARFF Header Section
      • The ARFF Data Section
    • Sparse ARFF files
    • Instance weights in ARFF files
  • XRFF
    • File extensions
    • Comparison
      • ARFF
      • XRFF
    • Sparse format
    • Compression
    • Useful features
      • Class attribute specification
      • Attribute weights
      • Instance weights
  • Converters
    • Introduction
    • Usage
      • File converters
      • Database converters
  • Stemmers
    • Introduction
    • Snowball stemmers
    • Using stemmers
      • Commandline
      • StringToWordVector
    • Adding new stemmers
  • Databases
    • Configuration files
    • Setup
    • Missing Datatypes
    • Stored Procedures
    • Troubleshooting
  • Windows databases
• IV Appendix
  • Research
    • Citing Weka
    • Paper references
  • Using the API
    • Option handling
    • Loading data
      • Loading data from files
      • Loading data from databases
    • Creating datasets in memory
      • Defining the format
      • Adding data
    • Generating artificial data
      • Generate ARFF file
      • Generate Instances
    • Randomizing data
    • Filtering
      • Batch filtering
      • Filtering on-the-fly
    • Classification
      • Building a classifier
      • Evaluating a classifier
      • Classifying instances
    • Clustering
      • Building a clusterer
      • Evaluating a clusterer
      • Clustering instances
    • Selecting attributes
      • Using the meta-classifier
      • Using the filter
      • Using the API directly
    • Saving data
      • Saving data to files
      • Saving data to databases
    • Visualization
      • ROC curves
      • Graphs
        • Tree
        • BayesNet
    • Serialization
  • Extending WEKA
    • Writing a new Classifier
      • Choosing the base class
      • Additional interfaces
      • Packages
      • Implementation
        • Methods
        • Guidelines
    • Writing a new Filter
      • Default approach
        • Implementation
        • Examples
      • Simple approach
        • SimpleBatchFilter
        • SimpleStreamFilter
        • Internals
      • Capabilities
      • Packages
      • Revisions
      • Testing
        • Option handling
        • GenericObjectEditor
        • Source code
        • Unit tests
    • Writing other algorithms
      • Clusterers
      • Attribute selection
      • Associators
    • Extending the Explorer
      • Adding tabs
        • Requirements
        • Examples
      • Adding visualization plugins
        • Introduction
        • Predictions
        • Errors
        • Graphs
        • Trees
    • Extending the Knowledge Flow
      • Creating a simple batch processing Step
      • Creating a simple streaming Step
      • Features of StepManager
      • PairedDataHelper
  • Weka Packages
    • Where does Weka store packages and other configuration stuff?
    • Anatomy of a package
      • The description file
      • Additional configuration files
    • Contributing a package
    • Creating a mirror of the package meta data repository
  • Technical documentation
    • ANT
      • Basics
      • Weka and ANT
    • CLASSPATH
      • Setting the CLASSPATH
      • RunWeka.bat
      • java -jar
    • Subversion
      • General
      • Source code
      • JUnit
      • Specific version
      • Clients
    • GenericObjectEditor
      • Introduction
      • File Structure
      • Exclusion
      • Class Discovery
      • Multiple Class Hierarchies
      • Capabilities
    • Properties
      • Precedence
      • Examples
    • XML
      • Command Line
      • Serialization of Experiments
      • Serialization of Classifiers
      • Bayesian Networks
      • XRFF files
  • Other resources
    • Mailing list
    • Troubleshooting
      • Weka download problems
      • OutOfMemoryException
        • Windows
      • Mac OSX
      • StackOverflowError
      • just-in-time (JIT) compiler
      • CSV file conversion
      • ARFF file doesn't load
      • Spaces in labels of ARFF files
      • CLASSPATH problems
      • Instance ID
        • Adding the ID
        • Removing the ID
      • Visualization
      • Memory consumption and Garbage collector
      • GUIChooser starts but not Experimenter or Explorer
      • KnowledgeFlow toolbars are empty
      • Links
  • Bibliography