~/talk

RICO: A Mobile App Dataset for Building Data-Driven Design Applications

on YouTube 16m
Talkat UIST 2017 · 2017-10-22..10-25· [data-visualization, ux-and-product-design]

speakers

name:Biplab Deka
links:none on file
name:Zifeng Huang
links:none on file
name:Chad Franzen
links:none on file
name:Joshua Hibschman
links:none on file
name:Daniel Afergan
links:none on file
name:Yang Li
links:none on file
name:Jeffrey Nichols
links:none on file
name:Ranjitha Kumar
links:none on file

resources

video:https://youtu.be/bVKgsNazl7w
slides:none on file

summary

A public dataset collects interaction traces and view hierarchies from 9.7k apps totaling 72k screens and 3M UI elements with over 10k recorded traces, built by combining 13 crowd workers (≈2,500 hours, ≈$2 per trace) and automated crawling. Screenshots, developer IDs, and view trees enable deduplicating designs and training an autoencoder that produces 64-dim layout vectors for query-by-example similarity search; example retrievals return coherent groups like login pages, galleries, and date pickers. Data and explorer are available at interactionmining.org/RICO.

AI-written summary · about

chapters (10)

  1. 0:00Introduction and collaborators
  2. 0:25RICO dataset overview
  3. 1:06Exploring RICO dataset online
  4. 4:08Hybrid mining approach
  5. 6:55Crowd worker challenges
  6. 7:16UI similarity annotations
  7. 8:40Layout vector training
  8. 10:44Future data enrichments
  9. 12:10Data-driven design applications
  10. 14:35Q&A on dark patterns

related talks

~/search
type to search 6,249 conferences, 138,987 speakers and 185,172 talks
~/help
cmdctrl k
search every conference, speaker and series
/
focus the filter search
h
home
c
conferences
s
speakers
e
series
t
light / dark — the status bar button does the same
?
this help
esc
close this, then clear every filter — or just the search, while you are typing in it

Ignored while you are typing. Pages without a filter rail have no search to focus.