The MineSet Tutorial introduces MineSet, an integrated suite of data mining and visualization tools, and provides a swift survey of the concepts and processes of data mining. This tutorial describes a few basic tasks to help you use MineSet immediately. Once you are familiar with the interface, refer to the MineSet User's Guide for a full description of other MineSet features. The Guide is delivered online as part of the MineSet product. See also http://mineset.sgi.com for more information.
This tutorial is for end users. No experience in programming is required, nor is any previous knowledge of statistics, although a basic knowledge of UNIX is assumed.
To work with this tutorial, MineSet should be installed on your system, or you should have access to such a system. The examples depend on it. Instructions for installing MineSet are available in the MineSet User's Guide and on the MineSet Web page http://mineset.sgi.com , where MineSet itself can be downloaded for evaluation purposes.
For this tutorial you do not need access to a database. The data needed is included in the MineSet distribution.
Chapter 1, “Data Mining Fundamentals,” introduces the concept of data mining and explains how it can be used to solve problems. Common data mining tasks are aligned with the various MineSet tools, although details are covered in later chapters.
Chapter 2, “Data Mining Process,” describes the tasks involved in the process of data mining. A case study of data mining using MineSet is provided.
Chapter 3, “Churn Tutorial,” provides a detailed tutorial for the process of data mining using MineSet. It begins from the initial screen and steps screen by screen through using MineSet tools on churn, a dataset provided with the MineSet distribution.
Chapter 4, “Further Explorations,” continues the exploration of MineSet with more complex variations of exploring data mining.
This tutorial uses several font conventions:
| italics | Italics are used for command and reference page names, filenames, variables, hostnames, user IDs, and the first use of new terms. | |
| Courier | Courier is used for examples of system output and for the contents of files. | |
| Courier bold | Courier bold is used for commands and other text that you are to type literally. |