Visualizing Association Rules in Groceries Yuqing Yang CS

  • Slides: 15
Download presentation
Visualizing Association Rules in Groceries Yuqing Yang CS 548 Showcase Prof. Carolina Ruiz

Visualizing Association Rules in Groceries Yuqing Yang CS 548 Showcase Prof. Carolina Ruiz

References � [1] Hahsler M, Chelluboina S. Visualizing Association Rules: Introduction to the R-extension

References � [1] Hahsler M, Chelluboina S. Visualizing Association Rules: Introduction to the R-extension Package arules. Viz[J]. R project module, 2011. � [2] Wong P C, Whitney P, Thomas J. Visualizing association rules for text mining[C]//Information Visualization, 1999. (Info Vis' 99) Proceedings. 1999 IEEE Symposium on. IEEE, 1999: 120 -123, 152. � [3] Ertek, Gürdal, and Ayhan Demiriz. "A framework for visualizing association mining results. " Computer and Information Sciences–ISCIS 2006. Springer Berlin Heidelberg, 2006. 593 -602. � [4] Hofmann, Heike, Arno PJM Siebes, and Adalbert FX Wilhelm. "Visualizing association rules with interactive mosaic plots. " Proceedings of the sixth ACM SIGKDD international conference on Knowledge discovery and data mining. ACM, 2000. � [5] Jeffrey Heer, Stuart K. Card, James Landay (2005). "Prefuse: a toolkit for interactive information visualization". In: ACM Human Factors in Computing Systems CHI 2005. 4/29/2014 2

1. Scatter Plot �Axes: two interest measures. �Color (gray level) – a third measure.

1. Scatter Plot �Axes: two interest measures. �Color (gray level) – a third measure. Figure 1: Scatter Plot[1] 4/29/2014 3

Two-key plot �Color – Order, the number of items contained in the rule. Figure

Two-key plot �Color – Order, the number of items contained in the rule. Figure 2: Scatter Plot[1] 4/29/2014 4

2. Matrix-based Visualizations Figure 3: Matrix-based visualization of two measures with colored squares [1]

2. Matrix-based Visualizations Figure 3: Matrix-based visualization of two measures with colored squares [1] Figure 4: Matrix-based visualization of two measures with colored squares (reordered)[1] 4/29/2014 5

Matrix-based Visualizations � 4/29/2014 6

Matrix-based Visualizations � 4/29/2014 6

Matrix-based Visualizations 3 D Bars Figure 5: Matrix-based visualization with 3 D bars[1] 4/29/2014

Matrix-based Visualizations 3 D Bars Figure 5: Matrix-based visualization with 3 D bars[1] 4/29/2014 7

3. Grouped Matrix-based Visualization �Columns -- antecedent groups �Rows – consequents �Color – aggregated

3. Grouped Matrix-based Visualization �Columns -- antecedent groups �Rows – consequents �Color – aggregated interest measure �Size of ballo 0 n -- aggregated support Figure 6: Grouped matrix with k=? [1] 4/29/2014 8

4/29/2014 9

4/29/2014 9

4. Graph-based Visualizations Figure 8: Itemsets as vertices[1] Figure 9: Rules as vertices[1] 4/29/2014

4. Graph-based Visualizations Figure 8: Itemsets as vertices[1] Figure 9: Rules as vertices[1] 4/29/2014 10

5. Parallel Coordinates Plot Figure 10: Parallel coordinate plot (reordered)[1] 4/29/2014 11

5. Parallel Coordinates Plot Figure 10: Parallel coordinate plot (reordered)[1] 4/29/2014 11

Parallel Coordinates Plot �To visualize multidimensional data. �X-axis -- the positions in a rule,

Parallel Coordinates Plot �To visualize multidimensional data. �X-axis -- the positions in a rule, i. e. , first item, second item, etc. �Head of arrow -- points to the consequent item. �The width of the arrows -- support �The intensity of the color -- confidence. 4/29/2014 12

Visualizing AR in Text Mining �A visualization of item associations with support > 0.

Visualizing AR in Text Mining �A visualization of item associations with support > 0. 4% and confidence > 50%. Figure 11: A visualization of item associations)[2] 4/29/2014 13

Infovis application Framework a toolkit for interactive information visualization � Provides theoretically-motivated abstractions for

Infovis application Framework a toolkit for interactive information visualization � Provides theoretically-motivated abstractions for the design of a wide range of visualization applications, enabling programmers to string together desired components quickly to create and customize working visualizations[5] � E. g. racialgraphic � Prefuse: � Ajax. org (Javascript) � Any. Chart (Flash) � Axiis � Degrafa � Ext. Js 4/29/2014 14

Inforvis Software and tools �Software � Panopticon http: //www. panopticon. com/ � Circos (Perl)

Inforvis Software and tools �Software � Panopticon http: //www. panopticon. com/ � Circos (Perl) http: //circos. ca/ � Balsamiq (hand-draw style) http: //webdemo. balsamiq. com/ �Web infovis � Easel. ly (story telling) http: //www. easel. ly/ � Piktochart http: //piktochart. com/ � Visual. ly http: //visual. ly/ � Infogr. am http: //infogr. am/ 4/29/2014 15