2:20 PM - 2:40 PM
[1M4-OS-20b-01] Fundamental Study of Hierarchical Graph Layout with Deep Learning
Keywords:Graph, Layout
Graphs are visualization methods that express the relationships between individual data with nodes and edges and have been used in various fields. Although graphs are a useful visualization method, the layout of the nodes greatly affects the readability. In particular, a hierarchical graph composed of a set of multiple nodes is known as an effective method to visualize the outline of a large-scale graph. It is very difficult to generate a highly readable layout because of complication of connection and the necessity of considering meta information.
Many methods have been studied for graph layout generation, but it is not always able to realize a flexible layout. Recently, graph layout methods using deep learning have been actively discussed, but layout generation of hierarchical graphs has not yet been achieved.
In this report, we focus on improving the layout of hierarchical graphs and propose a layout generation model using deep learning. We will work on improving the layout of hierarchical graphs with a deep learning model that uses highly evaluated graphs as training data.
Many methods have been studied for graph layout generation, but it is not always able to realize a flexible layout. Recently, graph layout methods using deep learning have been actively discussed, but layout generation of hierarchical graphs has not yet been achieved.
In this report, we focus on improving the layout of hierarchical graphs and propose a layout generation model using deep learning. We will work on improving the layout of hierarchical graphs with a deep learning model that uses highly evaluated graphs as training data.
Authentication for paper PDF access
A password is required to view paper PDFs. If you are a registered participant, please log on the site from Participant Log In.
You could view the PDF with entering the PDF viewing password bellow.