Storywrangler

Overview

Storywrangler started out as a platform for tracking and analyzing n-grams across Twitter's massive data stream. It has evolved to become a set of tools for analyzing stories of all kinds in order to surface meaning in oceans of text. It has expanded to incorporate a variety of large-scale text corpora as ecological time series—living records of how collective attention shifts across populations and over time. These tools can act as a computational history of what happened and why it mattered.

Institute members involved

Chris Danforth

Core team

Peter Sheridan Dodds

Core team

Michael Arnold

Core team

Jonathan St-Onge

Core team

Ben Dexter Cooley

Core team

IC2S2 2026 Tutorial

This project is presented as a tutorial at the 2026 International Conference on Computational Social Science (IC2S2) in Burlington, Vermont.

The original Storywrangler

Explore trending words and phrases across 100+ languages from 2008 to the 2023

Wikiwrangler

Use rank-turbulence metrics and the Allotaxonometer to explore what grabs collective attention at the national level on Wikipedia.

Research Focus

• Large-scale n-gram analysis across social media
• Multilingual trend detection and comparison
• Temporal dynamics of language and narratives
• Cross-cultural information diffusion
• Data visualization for massive text corpora