Browsing hierarchical data with multilayer dynamic queries and pruning

1
Browsing hierarchical data with multi-layer
dynamic queries and pruning

• Presented by Seth Hettich

2
Goals

• Enable users to browse hierarchical data
• Browse and view the data at all levels
• Query nodes in the tree
• Hide and expand sub-trees quickly
• Provide a high level overview of the data

3
PDQ Tree Features

• Two coupled views of the data
• Customizable dynamic query tools
• Dynamic queries at different levels of tree
• Pruning sub-trees to save space

4
PDQ Tree UI

• Data Display
• Query panel
• Overview

5
Using the query tool
6
Using the query tool
7
Data Display
8
Active nodes at current level
9
Active nodes at higher levels
10
Pruned nodes
11
PDQ Tree limits

• Need to tune UI for tree depth
• Only three attributes per level for query
• Only AND queries
• Not the most compact layout
• Limited user interactivity (cant click on nodes)

12
User study

• Tested three options
• Full-tree
• Partially-pruned
• Fully-pruned
• 24 subjects, 8 per option
• Timed at tasks, user evaluation

13
User Study results

• Fully-pruned was the fastest
• Partial pruning was the most popular
• Full tree sucked

14
(No Transcript)