CMU-CS-24-156
Computer Science Department
School of Computer Science, Carnegie Mellon University



CMU-CS-24-156

WaybackUI: A Dataset to Support the
Longitudinal Analysis of User Interfaces

Xin Yue Amanda Li

M.S. Thesis

December 2024

CMU-CS-24-156.pdf


Keywords: Human Computer Interaction, user interface, dataset, longitudinal analysis, design, accessibility, modeling

How do user interfaces (UIs) change over time? Understanding the evolution of UIs is essential for assessing the impact on users and automated systems that interact with them. To this end, we collected WaybackUI, a dataset of nearly 100,000 UIs from 2014 – 2024 mined from publicly available internet archives, paired with rendered screenshots and programmatically-extracted semantics. The resulting data allowed us to analyze how a decade of UI changes has impacted i) visual design, ii) accessibility, and iii) automated systems that interact with UIs. Examples of our findings include: i) a growing preference for muted color palettes in visual design, ii) an increase in the number of highly inaccessible web pages in recent years, and iii) quantifying the impact of data drift on the performance of UI understanding models. We conclude with a discussion of how WaybackUI can enable future data-driven discovery and understanding of UI trends.

49 pages

Thesis Committee:
Jeffrey Bigham (Chair)
Nik Martelaro

Srinivasan Seshan, Head, Computer Science Department
Martial Hebert, Dean, School of Computer Science


Return to: SCS Technical Report Collection
School of Computer Science

This page maintained by [email protected]