Reproduce “Detecting Narrative Shifts through Persistent Structures: A Topological Analysis of Media Discourse” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Mark M. Bailey, Mark I. Heiligman

Researchers from the National Intelligence University developed an unsupervised method using Topological Data Analysis (Persistent Homology) to detect and characterize structural shifts in media discourse following global events. The approach quantifies changes in narrative coherence (H0) and thematic framing (H1), demonstrating its ability to identify distinct patterns like 'bottom-up' or 'top-down' narrative reconfigurations across events such as the Russian invasion of Ukraine and the U.S. Capitol insurrection.

Key insights

  • Globally significant events induce rapid and profound transformations in the topological structure of public discourse, not just lexical changes.
  • The framework can distinguish between 'bottom-up' semantic reconfigurations (where H0 changes precede H1) and 'top-down' narrative imposition (where H1 changes lead H0), as observed during the Ukraine invasion.
  • Persistent Homology, applied to graphs of co-occurring noun phrases embedded in a metric space, offers a mathematically grounded, unsupervised, and scale-invariant tool for detecting and characterizing these structural shifts.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.