How It Works

From raw RSS feeds to a curated, classified, and connected news experience — here's the pipeline behind Newphoria.

The Pipeline

Newphoria continuously ingests articles from hundreds of RSS feeds spanning major wire services, scientific journals, space agencies, community networks, and independent publishers. Each article passes through a multi-stage AI pipeline:

• Ingestion — RSS feeds are polled, deduplicated, and normalized into a standard document format.
• AI Classification — The Coaptō engine reads each article and assigns a Bloom Score, category, and sentiment analysis.
• Entity Extraction — People, organizations, locations, and concepts are identified and linked.
• Knowledge Graph — Extracted entities and relationships are stored in a Neo4j graph database, connecting stories across time and topic.
• Static Site Generation — Curated articles are rendered into the static HTML you're reading now, deployed via Netlify.

The Bloom Score

Every article is scored on a four-category system we call the Bloom Score:

• GOOD — Constructive, solution-oriented, or progress-focused stories. Innovation, breakthroughs, acts of cooperation.
• BAD — Stories centered on conflict, decline, or harm. These are filtered from the main feed but tracked for framing analysis.
• UGLY — Sensationalism, outrage bait, or fear-driven coverage. Filtered out entirely.
• WEIRD — Genuinely strange and fascinating stories that defy easy classification. Unexplained phenomena, bizarre science, quantum oddities. These appear in The Wonderfully Weird section.

The score is generated by AI and reflects constructive value, not journalistic quality. A well-written article about a catastrophe may still score BAD because the subject matter is destructive rather than constructive.

Multi-Source Framing Analysis

When the same event is covered by multiple sources, Newphoria compares the framing across outlets. The same story can be presented as progress by one publisher and as a threat by another. Our AI identifies these framing differences and surfaces them, so you can see how the same facts get shaped by editorial perspective.

The Divergence Report

The Divergence Report is a periodic analysis of how coverage of the same event differs across sources. It highlights where outlets agree on facts but diverge on framing, emphasis, and tone. The goal is transparency — not to declare who is right, but to show you the full spectrum of how a story is being told.

The Technology Stack

Newphoria is powered by the Coaptō document intelligence engine, Neo4j for the knowledge graph, and a static site generator for the front end. The same AI platform that processes contracts and invoices for enterprise clients is what reads and classifies news for Newphoria — interpret unstructured content, classify it, and connect it.