Hello, I'm Shelly

I'm passionate about building world-class data analytics and visualization products.

Currently, I'm a data visualization developer at The Associated Press' Elections & Democracy team, where I design and build interactive visualizations for the 2024 presidential elections in the U.S.

My past work experience spans both tech and journalism industries. Most recently, I worked as a software engineer at Splunk, specializing in data visualization. I build interactive visual tools for businesses to explore and monitor large data infrastructure.

I have also worked as a graphics reporter/data visualization developer at The Wall Street Journal, NBC, NPR, and The Texas Tribune. I collaborated with reporters and editors to generate ideas, request and analyze public records to tell data-driven, visual stories. I covered stories on various topics, including high-profile companies in business, Texas local politics, and the impact of COVID-19 on employment.

My research paper on how captions affect visualization reading has been presented at IEEE Visualization Conference 2022 .

My visualization work has won awards from the European Journalism Centre, The Society of American Business Editors and Writers, The Society for News Design and Texas Medical Association, among others.


M.S. Computer Science, Columbia University
M.S. Journalism, Columbia University

B.A. Political Science, Colorado College

San Francisco, CA

Research

  • How Do Captions Affect Visualization Reading?

    Authors: [ Hazel Zhu, Shelly Cheng ]*, Eugene Wu (*equal contribution)
    IEEE Vis VisComm 2022

    Abstract: Captions help readers better understand visualizations. However, if the visualization is intended to communicate specific features, should the caption be statistical, and focus on specific values, or perceptual, and focus on general patterns? Prior work has shown that when captions mention visually salient features, users tend to recall those features. Still, we lack explicit guidelines for how to compose the appropriate caption. Further, what if the author wishes to emphasize a less salient feature?

Software Engineering

Front-end: HTML, CSS, JavaScript, TypeScript, D3.js, React.js, Styled-components, Maplibre-gl.js
Back-end: Python, Node.js, Flask, Java, Javalin, Linux
Testing: React testing library, Jest, Enzyme, Postman


Data Visualization

Data scraped with Selenium and BeautifulSoup, cleaned with Pandas and Regex in Python,

design in Illustrator, Sketch and InVision, and built with HTML, CSS, JavaScript, React.js, D3.js, Bootstrap, etc.


Infographics

Data analysis using Python

Visualizing using HTML, CSS, d3.js, QGIS, Illustrator and Ai2html



Using Illustrator for print newspapers