A SAMPLE WALK-THROUGH

Data Mining by Roiger and Geatz is tutorial, learning-by-doing approach. The following screenshots and verbal descriptions are taken from Chapters 2 and Chapters 4 and demonstrate how this learining-by-doing approach gets students performing unsupervised cluttering.

A Five-Step Approach for Unsupervised Cluttering (taken from Fig. 4.4)

1. Enter data to be mined into a new Excel spreadsheet.
2. Perform a data mining session.
3. Read and interpret summary results.
4. Read and interpret results for individual clusters.
5. Visualize and interpret rules defining the individual clusters.

STEP 1: Enter the data to be mined:

STEP 2: Perform a Data Mining Session:

(select parameter settings)

(generate rules)

STEP 3: Read and Interpret Summary Results:

(Sheet1 Res Sum)

(Sheet1 Res Sum)

STEP 4: Read and Interpret Individual Class Results:

(sheet1 res CLS)

(sheet1 res CLS)

STEP 5: Visualize Individual Class Rules:

(sheet1 res Rul)