C3Subtitles: 30C3: Data Mining for Good
back

Data Mining for Good

Using random sampling, entity resolution, communications metadata, and statistical modeling to assist prosecutions for disappearance and genocide in Guatemala

If you suspend your transcription on amara.org, please add a timestamp below to indicate how far you progressed! This will help others to resume your work!

Please do not press “publish” on amara.org to save your progress, use “save draft” instead. Only press “publish” when you're done with quality control.

Video duration
00:26:47
Language
English
Abstract
For over thirty years, human rights groups in Guatemala have carefully documented the killing and disappearance of many people in the early 1980s. There are tens of thousands of records in many databases, and over 80 million paper pages of police records available in the Archives of the National Police. Most of the prosecutions of the former military and police officials who committed the atrocities depends on eyewitnesses, specific documents, and forensic anthropologists' examination of exhumed bones. However, data analysis helps to see the big patterns in the violence.

This talk will explain how data analysis illuminated the selective patterns among mass killings in the prosecution for genocide of former de facto President General José Efraín Ríos Montt. The talk will also explain how looking at the communications metadata from over 20,000 randomly sampled paper memos helped illuminate command patterns in a disappearance case.

Talk ID
5405
Event:
30C3
Day
3
Room
Saal G
Start
5:30 p.m.
Duration
00:30:00
Track
Ethics, Society & Politics
Type of
lecture
Speaker
Patrick

Talk & Speaker speed statistics

Very rough underestimation:
164.1 wpm
908.1 spm
167.3 wpm
925.4 spm
100.0% Checking done100.0%
0.0% Syncing done0.0%
0.0% Transcribing done0.0%
0.0% Nothing done yet0.0%

Talk & Speaker speed statistics with word clouds

Whole talk:
164.1 wpm
908.1 spm
Patrick:
167.3 wpm
925.4 spm