Evaluating the Readability of Force Directed Graph Layouts: A Deep Learning Approach

@article{Haleem2019EvaluatingTR,
  title={Evaluating the Readability of Force Directed Graph Layouts: A Deep Learning Approach},
  author={Hammad Haleem and Y. Wang and Abishek Puri and Sahil Wadhwa and Huamin Qu},
  journal={IEEE Computer Graphics and Applications},
  year={2019},
  volume={39},
  pages={40-53}
}
Existing graph layout algorithms are usually not able to optimize all the aesthetic properties desired in a graph layout. [] Key Method A convolutional neural network architecture is proposed and trained on a benchmark dataset of graph images, which is composed of synthetically generated graphs and graphs created by sampling from real large networks. Multiple representative readability metrics (including edge crossing, node spread, and group overlap) are considered in the proposed approach. We quantitatively…

Figures and Tables from this paper

DeepGD: A Deep Learning Framework for Graph Drawing Using GNN

A Convolutional Graph Neural Network based deep learning framework, DeepGD, which can draw arbitrary graphs once trained, which attempts to generate layouts by compromising among multiple pre-specified aesthetics considering a good graph layout usually complies with multiple aesthetics simultaneously.

DeepDrawing: A Deep Learning Approach to Graph Drawing

This work proposes using a graph-LSTM-based approach to directly map network structures to graph drawings and evaluates the proposed approach on two special types of layouts and two general type of layouts in both qualitative and quantitative ways.

A Deep Generative Model for Graph Layout

  • Oh-Hyun KwonK. Ma
  • Computer Science
    IEEE Transactions on Visualization and Computer Graphics
  • 2020
This paper designs an encoder-decoder architecture to learn a model from a collection of example layouts, where the encoder represents training examples in a latent space and the decoder produces layouts from the latent space.

SmartGD: A Self-Challenging Generative Adversarial Network for Graph Drawing

A student-teacher GAN-based graph drawing framework, SmartGD, which learns to draw graphs just like how humans learn to perform tasks, to learn the optimal layout distribution with respect to any aesthetic criterion, whether the criterion is differentiable or not.

A Machine Learning Approach for Predicting Human Preference for Graph Layouts*

This paper presents the first machine learning approach for predicting qualitative evaluation of graph layouts using human preference experiment data, and to the best knowledge, this is the first approach.

Graph Exploration with Embedding-Guided Layouts

An embedding-driven graph layout algorithm is presented, GEGraph, which can achieve aesthetic layouts with better community preservation to support an easy interpretation of the graph structure and leverage embedding algorithms for attributed graphs to encode the two perspectives into latent space.

Aesthetic Discrimination of Graph Layouts

A neural network-based discriminator model trained on a labeled dataset that decides which of two layouts has a higher aesthetic quality is proposed, outperforming discriminators based on stress and on the linear combination of popular quality metrics.

Interactive Visual Pattern Search on Graph Data via Graph Representation Learning

A novel GNN for node-alignment called NeuroAlign is proposed, to facilitate easy validation and interpretation of the query results of graph neural networks used to encode a graph as fixed-length latent vector representation and perform subgraph matching in the latent space.

The Sprawlter Graph Readability Metric: Combining Sprawl and Area-Aware Clutter

An area-aware approach to clutter metrics is proposed that tracks the extent of geometric overlaps between node-node, node-edge, and edge-edge pairs in detail and is called a sprawlter metric, which handles variable-size nodes and explicitly treats metanodes and leaf nodes uniformly.

A Survey on ML4VIS: Applying MachineLearning Advances to Data Visualization.

This survey reveals seven main processes where the employment of ML techniques can benefit visualizations, related to existing visualization theoretical models in an ML4VIS pipeline, aiming to illuminate the role of ML-assisted visualization in general visualizations.

References

SHOWING 1-10 OF 42 REFERENCES

Aesthetic Discrimination of Graph Layouts

A neural network-based discriminator model trained on a labeled dataset that decides which of two layouts has a higher aesthetic quality is proposed, outperforming discriminators based on stress and on the linear combination of popular quality metrics.

What Would a Graph Look Like in this Layout? A Machine Learning Approach to Large Graph Visualization

A machine learning approach to large graph visualization based on computing the topological similarity of graphs using graph kernels is presented and an important contribution of this work is the development of a new framework to design graph kernels.

Visualizing Dataflow Graphs of Deep Learning Models in TensorFlow

Overall, users find the TensorFlow Graph Visualizer useful for understanding, debugging, and sharing the structures of their models.

Improving Graph Drawing Readability by Incorporating Readability Metrics : A Software Tool for Network Analysts

These contributions are aimed at heightening the awareness of network analysts that the images they share or publish could be of higher quality, so that readers could extract relevant information.

Towards Better Analysis of Deep Convolutional Neural Networks

A hybrid visualization is developed to disclose the multiple facets of each neuron and the interactions between them and a biclustering-based edge bundling method is proposed to reduce visual clutter caused by a large number of connections between neurons.

Graph Layouts by t‐SNE

A new graph layout method, tsNET, based on representing a graph with a distance matrix, which together with a modified t‐SNE cost function results in desirable layouts, which is evaluated by a formal comparison with state‐of‐the‐art methods.

AmbiguityVis: Visualization of Ambiguity in Graph Layouts

Metrics, including newly proposed metrics for abnormal edge lengths, visual overlap in community structures and node/edge aggregation, are proposed to quantify areas of ambiguity in the drawing and displayed using a heatmap-based visualization that provides visual feedback to developers of graph drawing and visualization approaches, allowing them to quickly identify misleading areas.

Readability metric feedback for aiding node-link visualization designers

This work detail new global readability metrics on a "0,1] continuous scale for node-node overlap, edge crossing angle, angular resolution, group overlap, and visualization coverage, and describes the trade-offs inherent in optimizing individual metrics as well as recommend metric optimizations for particular tasks.

A Treemap Based Method for Rapid Layout of Large Graphs

A new approach to the large graph layout problem is presented, which quickly generates an effective layout and can be used as the input to an iterative algorithm (e.g., a force directed method), which greatly reduces the number of iterations required to converge to a near optimal layout.

Perceptual Organization in User-Generated Graph Layouts

It is found that observers are able to recover cluster structure, that the distance between clusters is inversely related to the strength of the clustering, and that users exhibit the tendency to use edges to visually delineate perceptual groups.