An Excel-based Data Mining Tool

1. An Excel-based Data Mining Tool Chapter 4

2. 4.1 The iData Analyzer

3. Figure 4.1 The iDA system architecture

4. Figure 4.2 A successful installation

5. 4.2 ESX: A Multipurpose Tool for Data Mining Figure 4.3 An ESX concept hierarchy

6. 4.3 iDAV Format for Data Mining

8. 4.4 A Five-step Approach for Unsupervised Clustering Step 1: Enter the Data to be Mined Step 2: Perform a Data Mining Session Step 3: Read and Interpret Summary Results Step 4: Read and Interpret Individual Class Results Step 5: Visualize Individual Class Rules

9. Step 1: Enter The Data To Be Mined Figure 4.4 The Credit Card Promotion Database

10. Step 2: Perform A Data Mining Session Figure 4.5 Unsupervised settings for ESX

11. Figure 4.6 RuleMaker options

12. Step 3: Read and Interpret Summary Results Class Resemblance Scores Domain Resemblance Score Domain Predictability

13. Figure 4.8 Summery statistics for the Acme credit card promotion database

14. Figure 4.9 Statistics for numerical attributes and common categorical attribute values

15. Step 4: Read and Interpret Individual Class Results Class Predictability is a within-class measure. Class Predictiveness is a between- class measure.

16. Figure 4.10 Class 3 summary results

17. Figure 4.11 Necessary and sufficient attribute values for Class 3

18. Step 5: Visualize Individual Class Rules Figure 4.7 Rules for the credit card promotion database

19. 4.5 A Six-Step Approach for Supervised Learning Step 1: Choose an Output Attribute Step 2: Perform the Mining Session Step 3: Read and Interpret Summary Results Step 4: Read and Interpret Test Set Results Step 5: Read and Interpret Class Results Step 6: Visualize and Interpret Class Rules

20. Read and Interpret Test Set Results Figure 4.12 Test set instance classification

21. 4.6 Techniques for Generating Rules • Simple Procedure for Creating Best Set of Covering Rules • Choose an attribute that best differentiate all domains. • Use the attribute to further subdivide instances into classes. • For each subclass created in step 2 • 3.1 If the instances in the subclass satisfy a predefined criteria • Then generate a defining rule for the subclass. • 3.2 If the subclass does not satisfy the predefined criteria • Then repeat step 1

22. 4.6 Techniques for Generating Rules (RuleMaker) Define the scope of the rules. Choose the instances. Set the minimum rule correctness. Define the minimum rule coverage. Choose an attribute significance value.

23. 4.7 Instance Typicality • The average similarity of instance to all other instances within its class. • Identify prototypical and outlier instances. • Select a best set of training instances. • Used to compute individual instance classification confidence scores.

24. Figure 4.13 Instance typicality

25. 4.8 Special Considerations and Features Avoid Mining Delays – at some point copy the original data into another Excel sheet The Quick Mine Feature – recommended when the dataset contains more than 2000 instances Erroneous and Missing Data – blank lines, beyond the last column, invalid characters