Data2Vis: Automatic Generation of Data Visualizations Using Sequence-to-Sequence Recurrent Neural Networks

@article{Dibia2019Data2VisAG,
  title={Data2Vis: Automatic Generation of Data Visualizations Using Sequence-to-Sequence Recurrent Neural Networks},
  author={Victor C. Dibia and Çagatay Demiralp},
  journal={IEEE Computer Graphics and Applications},
  year={2019},
  volume={39},
  pages={33-46}
}
Rapidly creating effective visualizations using expressive grammars is challenging for users who have limited time and limited skills in statistics and data visualization. Even high-level, dedicated visualization tools often require users to manually select among data attributes, decide which transformations to apply, and specify mappings between visual encoding variables and raw or transformed attributes. In this paper we introduce Data2Vis, an end-to-end trainable neural translation model for… 

Figures and Tables from this paper

PlotCoder: Hierarchical Decoding for Synthesizing Visualization Code in Programmatic Context
TLDR
This paper introduces PlotCoder, a new hierarchical encoder-decoder architecture that models both the code context and the input utterance and uses it to first determine the template of the visualization code, followed by predicting the data to be plotted.
Advancing Visual Specification of Code Requirements for Graphs
TLDR
This paper uses a hybrid model, combining a neural network and optical character recognition to generate the code to create the visualization, and allows the user to visually specify their code requirements in order to lower the barrier for humanities researchers to learn how to program visualizations.
Towards Natural Language Interfaces for Data Visualization: A Survey
TLDR
This article conducts a comprehensive review of the existing V-NLIs and develops categorical dimensions based on a classic information visualization pipeline with the extension of a V-NLI layer.
AI4VIS: Survey on Artificial Intelligence Approaches for Data Visualization.
  • Aoyu Wu, Yun Wang, Huamin Qu
  • Computer Science, Art
    IEEE transactions on visualization and computer graphics
  • 2021
TLDR
This survey probes the underlying vision of formalizing visualizations as an emerging data format and review the recent advance in applying AI techniques to visualization data (AI4VIS), and defines visualization data as the digital representations of visualizations in computers and focus on data visualization.
VizML: A Machine Learning Approach to Visualization Recommendation
TLDR
A novel machine learning-based approach to visualization recommendation that learns visualization design choices from a large corpus of datasets and associated visualizations that is comparable to human performance when predicting consensus visualization type and exceeds that of other visualization recommender systems.
Analyzing the Feasibility of Generating Data Visualizations from Hand-drawn Sketches Using Deep Learning
  • Computer Science
  • 2021
TLDR
This paper investigates the feasibility of using deep learning techniques and tools to generate the source code for multi-platform data visualizations automatically from hand-drawn sketches provided by a domain expert.
A Survey on ML4VIS: Applying MachineLearning Advances to Data Visualization.
TLDR
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.
Calliope: Automatic Visual Data Story Generation from a Spreadsheet
TLDR
This paper introduces a novel visual data story generating system, Calliope, which creates visual data stories from an input spreadsheet through an automatic process and facilities the easy revision of the generated story based on an online story editor.
Survey on Artificial Intelligence Approaches for Visualization Data
TLDR
This survey probes the underlying vision of formalizing visualizations as an emerging data format, and discusses several important research questions surrounding the management and exploitation of visualization data, as well as the role of AI in support of those processes.
Learning to Recommend Visualizations from Data
TLDR
This work proposes the first end-to-end ML-based visualization recommendation system that leverages a large corpus of datasets and their relevant visualizations to learn a visualization recommendation model automatically and recommends more effective and useful visualizations compared to existing state-of-the-art rule-based systems.
...
...

References

SHOWING 1-10 OF 97 REFERENCES
DeepEye: Creating Good Data Visualizations by Keyword Search
TLDR
DeepEye is an innovative visualization system that aims at helping everyone create good visualizations simply like a Google search, Given a dataset and a keyword query, DeepEye understands the query intent, generates and ranks goodVisualizations.
VizML: A Machine Learning Approach to Visualization Recommendation
TLDR
A novel machine learning-based approach to visualization recommendation that learns visualization design choices from a large corpus of datasets and associated visualizations that is comparable to human performance when predicting consensus visualization type and exceeds that of other visualization recommender systems.
Reverse‐Engineering Visualizations: Recovering Visual Encodings from Chart Images
TLDR
An end‐to‐end pipeline which takes a bitmap image as input and returns a visual encoding specification as output is contributed and accurate automatic inference of text elements, mark types, and chart specifications across a variety of input chart types is demonstrated.
Voyager: Exploratory Analysis via Faceted Browsing of Visualization Recommendations
TLDR
It is found that Voyager facilitates exploration of previously unseen data and leads to increased data variable coverage, and the need to balance rapid exploration and targeted question-answering for visualization tools is distill.
Voyager 2: Augmenting Visual Analysis with Partial View Specifications
TLDR
This work presents Voyager 2, a mixed-initiative system that blends manual and automated chart specification to help analysts engage in both open-ended exploration and targeted question answering and contributes two partial specification interfaces.
AutoVis: Automatic Visualization
TLDR
This work distinguishes an automatic visualization system (AVS) from an automated visualization system, a programming system for automating the production of charts, graphs and visualizations, designed to protect researchers from ignoring missing data, outliers, miscodes and other anomalies that can violate statistical assumptions or otherwise jeopardize the validity of models.
SeeDB: Efficient Data-Driven Visualization Recommendations to Support Visual Analytics
TLDR
This work proposes SeeDB, a visualization recommendation engine to facilitate fast visual analysis: given a subset of data to be studied, SeeDB intelligently explores the space of visualizations, evaluates promising visualizations for trends, and recommends those it deems most “useful” or “interesting”.
Latent Predictor Networks for Code Generation
TLDR
A novel neural network architecture is presented which generates an output sequence conditioned on an arbitrary number of input functions and allows both the choice of conditioning context and the granularity of generation, for example characters or tokens, to be marginalised, thus permitting scalable and effective training.
Seq2SQL: Generating Structured Queries from Natural Language using Reinforcement Learning
TLDR
This work proposes Seq2 SQL, a deep neural network for translating natural language questions to corresponding SQL queries, and releases WikiSQL, a dataset of 80654 hand-annotated examples of questions and SQL queries distributed across 24241 tables fromWikipedia that is an order of magnitude larger than comparable datasets.
Protovis: A Graphical Toolkit for Visualization
TLDR
Protovis, an extensible toolkit for constructing visualizations by composing simple graphical primitives, is contributed, which achieves a level of expressiveness comparable to low-level graphics systems, while improving efficiency and accessibility.
...
...