Ontology Visualization

1
Ontology Visualization
2
Outline
1. Why Ontology Visualization
2. How can we help you
3. Architecture
4. Implementation
5. Demo
3
Why Ontology Visualization ?
1

• Ontology are machine understandable and thus
  need some means of graphical visualization for
  humans to comprehend
• Visualization tools makes it easy to understand
  large and complex ontology

4
What lacking in current Visualization tool

• Not all tools can display OWL RDF files
• Most of the tools dont have efficient
  navigational techniques
• Provides on Zoom Pan
• Most of the tool can only display subclass
  relationships
• Not all tools can display OWL restrictions
• Most tools do not provide options for browsing
  ontology using search
• Most tools do not provide layouts to visualize

5
Good Visualization Tool Features

• It should be able to load large ontology
• It should be able to display RDF OWL files
• Navigational techniques can be used to move to
  certain region in the large ontology
• It should have efficient navigation technique
  like zoom pan, browsing option through search
  etc..

6
How can we help you
2

• This tool would be very useful for the end users
  who are interested in viewing specific regions in
  the ontology
• Data is loaded from files (OWL/RDF), then parsed
  and displayed in graph
• The user can search for a given class or instance
  display its neighborhood using default hops
• Browsing of the ontology can be done using the
  search

7
File Loading
2

• The user can load schema and instance and choose
  to view either the schema or instance file
• Multiple files can be loaded but only one file
  can be displayed

8
Loading an OWL File
9
Loading an RDF File
10
Configuration File

• Screen shot

11
Display
2

• Initially the graph is displayed using
• the default layout
• centered using the default search term
• And nodes being displayed are within the default
  number of hopsof the searched term.
• These default options are set in the configurator
  file.
• After the initial display the user can change
  what layout the tool uses and what was searched
  for along with the number of nodes to display,
  using the Search, Hops, and Layout options in the
  GUI.

12
Changing Layout
2

• Chose the layout menu option
• Select the desired layout.
• The available layouts include
• SpringEmbedded
• Sugiyama
• Circle
• Moen
• Radial Tree
• Ordered Tree. (screenshot)

13
Layouts
Circle Layout
Spring Layout
14
Layouts Continued
Radial Layout
Annealing Layout
15
Searching
2

• The tool uses a Breadth First Search to get the
  nodes when in the desired number of hops.
• To search, go to the Search box above the main
  view of the graph and Type in the Label of the
  node you are searching for and press OK.
• The graph will be updated with the node
  containing the label in the center and the
  neighborhood based on the default hops
  (screenshot)

16
Search
17
Class Hierarchy Property Hierarchy
2

• To the side of the graph is a panel containing
  tree views of the file the user is displaying.
• The tree view can either be of the Class
  Hierarchy or of the Properties of the Class.
• (screenshot)

18
View Panels
Class View
Properties View
19
Other features

• The user can view instances of class by clicking
  on given class
• Zooming technique is provided to visualize large
  graphs
• The tool has Zoom-in Zoom-out feature provided
• Provides different Layouts for graphs

20
Screenshots
2

• SHOW some screenshots

21
System Design
22
Architecture
3
Main Memory Data Structures
23
APIs Used

• We had to implement the entire project from
  scratch
• Our project used the following APIs
• JGraph
• JGraphAddons (Layouts)
• SemDis (Main Memory data structure)

24
About JGraph API
4

• JGraph is a very powerful open source Java graph
  drawing and layout component, which is used by 12
  of the top 30 Fortune 500 companies
• JGraph is a graph component written in 100 pure
  Java
• Features of JGraph include
• zooming, folding, undo, drag and drop
• JGraph is based on the Swing MVC pattern

25
JGraph Features
26
SemDis Api

• Created by a group of people from LSDIS lab
• Uses Jena API internally
• Api provides classes to load RDF or OWL (without
  restrictions) and creates main memory data
  Structures

27
Implementation

• Class Hierarchy
• Property Hierarchy
• Creation of JGraph objects to load OWL/RDF
  Schema/Instance files
• Developing Breadth First Search Algorithm to
  support search
• Implemented on JGraph

28

• Display of Instances for a given Class
• Configuration file setup
• Creation of Graphical User Interface

29
Demo
5

• Demo show

30

• The End