Why a 12-year-old forecasting paper has stood the test of time
A 12-year-old Amazon forecasting paper just won KDD's test-of-time award. Here's why ML models predicting civil unrest still matter.
Why it matters
Amazon Scholar's decade-old machine learning work on predictive forecasting has proven durable enough to win institutional recognition, signaling that foundational ML research on real-world prediction problems maintains long-term value and relevance to enterprise AI applications.
The key facts
9 to know2014 paper on forecasting civil unrest in Latin America
Coauthored by Amazon Scholar Aravind Srinivasan
Won test-of-time award at KDD 2025
12-year research durability benchmark
Focus on predictive modeling for geopolitical events
Paper published 2014, won test-of-time award at KDD 2025
Coauthor: Aravind Srinivasan, Amazon Scholar
Focus: Forecasting civil unrest in Latin America using machine learning
Venue: KDD (Knowledge Discovery and Data Mining) conference
Go to the source
Amazon Scienceamazon.science
Publisher excerpt: Amazon Scholar Aravind Srinivasan coauthored a 2014 paper about forecasting civil unrest in Latin America, which won a test-of-time award at KDD 2025.
