1. 1 Visualizing Multi-dimensional Clusters, Trends, and Outliers using Star Coordinates Author : Eser Kandogan Reporter : Tze Ho-Lin 2007/5/9 SIGKDD, 2001

2. 2 Outline Motivation Objectives Methodology: Star Coordinates Interaction techniques Evaluation Conclusion Personal Comments

3. 3 Motivation Real datasets contain typically more than three attributes of data, representing and making sense of multi-dimensional data has been challenging.

4. 4 Objectives The objective for this paper is to relieve the dimensionality curse on knowledge discovery through simple data representations that are derived from familiar and easy to understand lower dimensional representations.

5. 5 Methodology

6. 6 j: 資料點 i: 屬性

7. 7 Interaction techniques 1. Scaling Scaling 2. Rotation Rotation 3. Marking 4. Range Selection Range Selection 5. Histogram Histogram 6. Footprints Footprints 7. Sticks Sticks

8. 8 Evaluation

9. 9 Conclusion Star Coordinates, aims to let a representation of the higher dimensional space built on the well- known simple representations and also through dynamic interactions that allow users to discover trends, outliers, and clusters easily.

10. 10 Personal Comments Application  Data visualization Advantage  Simple & Easy to understand Disadvantage  The figures in this paper is rough.

11. 11 Scaling

12. 12 Rotation

13. 13 Range Selection

14. 14 Histogram

15. 15 Footprints

16. 16 Sticks

17. 17 Evaluation - Figure 12

18. 18 Evaluation - Figure 13

19. 19 Evaluation - Figure 14 Figure 14. Data point distribution after removing state, area code, phone number, and total minute and calls for day, evening, night, and international calls.

20. 20 Evaluation - Figure 15 Figure 15. Data point partitioned into four clusters based on international service plan and voice plan membership.

21. 21 Evaluation - Figure 16 Figure 16. Total day charge and number of customer service calls play the most significant role in churn for customers without international and voice mail service plans ‧.