Articles enter through configured source feeds and are normalized into a common article shape.
0FDI 000
Methodology
How Fracture turns coverage into a readable signal.
Fracture watches related coverage across sources, clusters articles into stories, then measures where framing, tone, and headline emphasis diverge.
Related articles are grouped into story clusters using topic, timing, and source overlap.
The system scores differences in headlines, sentiment, framing type, and structural emphasis.
Fracture starts with source collection.
The ingestion pipeline pulls articles from configured feeds or APIs, preserves the outlet identity, and stores publication timing, headline text, summary data, image metadata, and known source priors.
Story clusters are built from shared events.
Articles become more useful when they are compared against other coverage of the same event. Fracture groups related articles so readers can inspect the spread rather than a single isolated account.
FDI measures framing distance.
The Fracture Divergence Index summarizes measurable distance in headline sentiment, framing type entropy, entity emphasis, source selection variance, and structural divergence.
The score is a signal, not a verdict.
A high FDI score means coverage is farther apart across measurable dimensions. It does not automatically mean one outlet is wrong or another is right.
Fracture keeps the product focused on source comparison, framing distance, and readable context instead of performative certainty.