Paper Info
Title
Disentangled Representation of Data Distributions in Scatterplots.
Abstract
We present a data-driven approach to obtain a disentangled and interpretable representation that can characterize bivariate data distributions of scatterplots. We first collect tabular datasets from the Web and build a training corpus consisting of over one million scatterplot images. Then, we train a state-of-the-art disentangling model, β-variational autoencoder, to derive a disentangled representation of the scatterplot images. The main output of this work is a list of 32 representative features that can capture the underlying structures of bivariate data distributions. Through latent traversals, we seek for high-level semantics of the features and compare them to previous human-derived concepts such as scagnostics measures. Finally, using the 32 features as an input, we build a simple neural network to predict the perceptual distances between scatterplots that were previously scored by human annotators. We found Pearson’s correlation coefficient between the predicted and perceptual distances was above 0.75, which indicates the effectiveness of our representation in the quantitative characterization of scatterplots.
Year
DOI
Venue
2019
10.1109/VISUAL.2019.8933670
VIS
Keywords
Field
DocType
Training,Neural networks,Visualization,Data visualization,Correlation,Machine learning,Generative adversarial networks
Correlation coefficient,Bivariate data,Autoencoder,Pattern recognition,Computer science,Visualization,Theoretical computer science,Human-centered computing,Artificial intelligence,Artificial neural network,Perception,Semantics
Conference
ISBN
Citations 
PageRank 
978-1-7281-4941-7
2
0.34
References 
Authors
0
2
Name
Order
Citations
PageRank
Jaemin Jo1547.35
Jinwook Seo258652.89