MineSet^™User's Guide

Table of Contents

About This Guide

Audience for This Guide

Structure of This Document

Illustration in This Guide

Typographical Conventions

1. Getting Started

MineSet Tools Suite

Basic Tool Execution Scenario

2. Setting Up MineSet

Configuring the DataMover Server

Loading Sample Datasets

3. The Tool Manager

Overview

Starting the Tool Manager

Choosing a Data Source

Transforming the Data

Investigating the Data

Session Files

Pulldown Menus

The Tool Manager Options File

The Record Viewer

Color Options for the MineSet Visualizers

4. Using the Statistics Visualizer

Overview of the Statistics Visualizer

File Requirements

Starting the Statistics Visualizer

Working in the Statistics Visualizer's Main Window

Pulldown Menus

The Help Menu

Sample Data Files

5. Using the Tree Visualizer

Overview of Tree Visualizer

File Requirements

Starting the Tree Visualizer

Configuring the Tree Visualizer Using the Tool Manager

Working in the Tree Visualizer's Main Window

External Controls

Pulldown Menus

Null Handling in the Tree Visualizer

Sample Configuration and Data Files

6. Using the Map Visualizer

Overview of Map Visualizer

File Requirements

Starting the Map Visualizer

Configuring the Map Visualizer Using the Tool Manager

Working in the Map Visualizer's Main Window

External Main Window Controls

The Animation Control Panel

Pulldown Menus

Null Handling in the Map Visualizer

Sample Configuration and Data Files

7. Using the Scatter Visualizer

Overview of Scatter Visualizer

File Requirements

Starting the Scatter Visualizer

Configuring the Scatter Visualizer Using the Tool Manager

Working in the Scatter Visualizer's Main Window

External Controls

The Animation Control Panel

Pulldown Menus

Sample Configuration and Data Files

8. Using the Splat Visualizer

Overview of the Splat Visualizer

File Requirements

Starting the Splat Visualizer

Configuring the Splat Visualizer Using the Tool Manager

Working in the Splat Visualizer's Main Window

External Controls

The Animation Control Panel

Pulldown Menus

Sample Configuration and Data Files

9. Using the Rules Visualizer

Overview of Rules Visualizer

File Requirements

Starting the Rules Visualizer

Configuring the Rules Visualizer Using the Tool Manager

Working in the Rules Visualizer's Main Window

External Controls

Pulldown Menus

Sample Files

10. MineSet Inducers and Classifiers

Classifiers

Inducers

Training Set

Error Estimation

Inducer Modes in Tool Manager

Error Options for Inducers

The Status Window

Applying Models, Testing Models, and Fitting New Data

Special Options and Limitations

11. Inducing and Visualizing the Decision Tree Classifier

Overview

Inducing Decision Trees

File Requirements

Running the Decision Tree Inducer

Configuring the Decision Tree Inducer Using the Tool Manager

Working in the Tree Visualizer's Main Window

External Controls

Pulldown Menus

Sample Files

12. Inducing and Visualizing the Option Tree Classifier

Overview

Inducing Option Trees

File Requirements

Running the Option Tree Inducer

Configuring the Decision Tree Inducer Using the Tool Manager

Working in the Tree Visualizer's Main Window

Sample Files

13. Inducing and Visualizing the Evidence Classifier

Overview

Inducing Evidence Classifiers

File Requirements

Running the Evidence Inducer

Starting the Evidence Visualizer

Configuring the Evidence Inducer Using the Tool Manager

Working in the Evidence Visualizer's Panes

External Controls

Pulldown Menus

Sample Files

14. Inducing and Visualizing the Decision Table

Overview

Inducing Decision Tables

File Requirements

Running the Decision Table Inducer

Starting the Decision Table Visualizer

Configuring the Decision Table Inducer Using the Tool Manager

Working in the Decision Table Visualizer's Main Window

External Main Window Controls

Pulldown Menus

Sample Files

15. Inducing and Visualizing the Regression Tree

Overview

Running the Regression Tree Inducer

Configuring the Regression Tree Inducer Using the Tool Manager

Visualizing the Regression Tree

External Controls

Pulldown Menus

Sample Files

16. Inducing and Visualizing Clustering

Overview of Clustering

Using Clustering and the Cluster Visualizer

Evaluation of Clustering

Starting the Cluster Visualizer

File Requirements

Working in Cluster Visualizer Main Window

Pulldown Menus

Sample File

Alternative Visualization of Clustering

17. Column Importance

Finding Important Columns

Column Importance Notes

Column Importance and Relation to Classifiers

Sample File

18. Selection and Drill-Through

Multiple Selection

Drill-Through

19. File Exchange Between MineSet and SAS

Overview

Converting MineSet Data Files to SAS Data Sets

Converting SAS Data Sets Into MineSet Data Files

20. MineSet Web Extensions

Overview

MineSet Web Extension Files

MineSet Web Installation (Client)

MineSet Web Installation (Server)

MineSet mtr Files

MineSet Remote View

MineSet Web Extension Security-Related Issues

A. Flat File Support for MineSet

The Data File

The .schema File

Exceptions

B. Creating Data and Configuration Files for the Tree Visualizer

The Data File

The Configuration File

Keywords

Expressions

The Input Section

The Expression Section

The Hierarchy Section

The View Section

C. Creating Data, Configuration, Hierarchy, and GFX Files for the Map Visualizer

The Data File

The Configuration File

The Hierarchy File

The .gfx File

D. Creating Data and Configuration Files for the Scatter Visualizer

The Data File

The Configuration File

The Input Section

The Expressions Section

The View Section

E. Creating Data and Configuration Files for the Splat Visualizer

The Data File

The Configuration File

The Input Section

The View Section

F. Creating Data and Configuration Files for the Rules Visualizer

The Association Data Converter

Association Rules Generator

Rules Visualization

G. Format of the Evidence Visualizer's Data File

H. Creating Data and Configuration Files for the Decision Table Visualizer

Sample File

I. Command-Line Interface to MIndUtil: Analytical Data Mining Algorithms

MIndUtil Invocation and Options

General Options

Induction Modes

Estimate Error

Learning Curve

Clustering

Discretization

Column Importance and Auto Selection

Fit-Data

MineSet-to-MLC, MLC-to-MineSet

Visualize

J. Nulls in MineSet

Semantics of Nulls

Representation of Nulls

Operations on Nulls

Aggregations in the Presence of Nulls

Sort Order for Nulls

Bins and Arrays With Nulls

K. Further Reading and Acknowledgments

Further Reading

Acknowledgments

Index