~/talk

Efficient Similar-Object Search Algorithms for Terabytes of Data

on YouTube 24m
TalkinRussianat HighLoad++ 2015 · 2015-11-02..11-03· [data-engineering, machine-learning]

speaker

name:Evgeniy Zhurin
links:none on file

resources

video:https://youtu.be/cvd0NkhjK3Y
slides:none on file

summary

Segmento DMP stores hundreds of millions of anonymous internet user profiles and needs to cluster them and find fuzzy duplicates. The talk covers computing similarity metrics between profiles at terabyte scale and explains alternatives to brute-force comparisons, focusing on scalable approximate search algorithms and practical engineering choices used to make similarity search feasible on that data volume.

AI-written summary · about

chapters (8)

  1. 0:00Introduction
  2. 3:38Look-alike audience problem
  3. 6:25MinHash for compact representation
  4. 9:38Locality-Sensitive Hashing
  5. 13:53Parameter tuning and accuracy trade-off
  6. 14:13SimHash algorithm and use
  7. 17:48Tech stack and implementation
  8. 18:37Q&A

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.