~/talk

Trust, but Verify: Optimistic Visualizations of Approximate Queries for Exploring Big Data

on YouTube 20m
Talkat CHI 2017 · 2017-05-06..05-11· [data-visualization]

speakers

name:Dominik Moritz
links:none on file
name:Danyel Fisher
links:none on file
name:Bolin Ding
links:none on file
name:Chi Wang
links:none on file

resources

video:https://youtu.be/4ntnRzqiG7Y
slides:none on file

summary

Pangloss implements optimistic visualization by running fast approximate queries with Sample+Seek and launching exact SQL Server queries in the background, letting a visualization show values plus error bars and heatmap uncertainty. On a 70 GB, ~170M-flight dataset approximate results return in ~100 ms on one machine; exact results complete later and caught a single missed flight that the sample missed. Five lab experts and three Microsoft case teams report higher confidence and smoother exploration.

AI-written summary · about

chapters (9)

  1. 0:00Introduction and motivation
  2. 2:08Challenges with big data exploration
  3. 3:58Approximate query processing explained
  4. 6:28Problems with approximate results
  5. 7:52Introducing optimistic visualization
  6. 9:58Demo of Pangloss system
  7. 12:33User study findings
  8. 15:05Why optimistic visualization works
  9. 16:48Future work and closing

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.