FINANVIZ● GENERATED
2018 · IEEE Big Data · Research

FinanViz.

A market moves on stories before it moves on numbers — so watch the stories form.

Stack
NLPTopic modelingStreaming dataD3.js

Financial news arrives as an unrelenting stream. Buried in it are topics that swell, peak, and fade — a regulatory rumor, a product launch, a crisis. FinanViz mines that stream for emerging themes and lays them out as flowing bands of attention over time.

FINANVIZ● GENERATED
A theme-river layout: each band is a topic, its thickness the volume of coverage. New bands appear as topics emerge; the eye tracks which stories are gathering momentum.

From text to topics

Behind the visualization sits a text-mining pipeline that clusters incoming articles into coherent topics and tracks each cluster’s strength over successive windows. The visual layer’s job is to make the dynamics legible — not just “what topics exist” but “which ones are accelerating right now.”

Who it is for

  • Analysts scanning for narratives before they hit the price.
  • Researchers studying how financial attention concentrates and disperses.
  • Anyone drowning in a feed who needs the shape, not every headline.

FinanViz was presented at IEEE Big Data 2018, part of an early move from pure signal processing toward large-scale, text-driven visual analytics.