Browse Prior Art Database

A method and system for online dynamic graph visualization

IP.com Disclosure Number: IPCOM000176029D
Original Publication Date: 2008-Nov-03
Included in the Prior Art Database: 2008-Nov-03
Document File: 2 page(s) / 45K

Publishing Venue

IBM

Abstract

In this artical, we proposed a new dynamic graph visualization system and an efficent dynamic loading method for supporting online visualization. The system contains three major modules : the dynamic data loading model, increamental layout module and the animation module. The dynamic laoding module is the fundation of the wholse system. A new indexing method was proposed in this artical to accelerate the increamental data loading for supporting online dynamic graph visualization.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 49% of the total text.

Page 1 of 2

A method and system for online dynamic graph visualization

1. Background

          are one of the most important data structures which are wildly used for data modeling. For example, all of the social relationships, the internet and the telecom net can be represented by a network structure. The visualization techniques of networks/graphs have also been well studied over the past decades. Many visualization methods have been designed to show the graphs in the different application domains [1]. However, the traditional researches mainly focused on visualizing a static view of the graph which ignores the fact that most networks will keep on changing over times (known as the dynamic graph). Directly showing a static view may lose some important information which is hidden along the time dimension. Thus, visualizing the dynamic graph/network becomes one of the hottest topics in the areas of graph visualization and social network analysis. The latest research is the Sonia system [2]. It visualizes a graph snapshot (frame) at any time by using traditional layout algorithms and transits between the frames by leveraging animation technologies to keep users' mental map.

Although the dynamic graph visualization in Sonia system is quite impressive, it is not an efficient system for the dynamic graphs from real-life applications. Sonia system does not support dynamic data loading; it loads all data into memory at the beginning of the process. Thus it can't be used for large dataset. Furthermore it is not a complete online dynamic graph visualization system.

To solve the above problems and better visualize the dynamic graphs, in this disclosure, we proposed a new method and system for dynamic graph visualization.

2.Summary of invention

Networks/graphs

                                                    Figure 1 illustrates the architecture of the dynamic graph visualization system. It contains four major modules. With a given period and step interval, the system runs in an iteratively way controlled by the transition controller. At the beginning of the each iteration, the dynamic loading module loads the changed data of the next graph incrementally based on the structure of the current graph from the database. The graph will be re-laid out in the layout module when new data are loaded. Finally graph elements will be animated transit to the new layout positions in the animation module. This process repeats until the final graph in the graph series is reached.

The dynamic data loading module is the foundation of the whole system. In the proposed system, the dynamic data loading module is quite different from the traditional data loading component. It works in the following way: when the change amount is not big, only the change context, such as the newly added elements, the unchanged elements and the deleted elements, is loaded; otherwise, the static snapshot of the graph with the change...