Persistent homology, topological data analysis, and the grand timeline
The algebra that classifies surfaces now finds the shape of data. Sweep a scale across point clouds and images to see which holes last, then trace the field from Euler to today.
Pick a point cloud, then drag the ε slider or press Sweep ε to grow balls around the points and build the complex. Long bars in the barcode are real features; short bars are noise.
Algebraic topology has escaped the ivory tower. Persistent homology applies the machinery of chain complexes and boundary operators to real-world data — detecting clusters, loops, and voids in point clouds, images, and sensor networks. The same algebra that classifies surfaces now classifies the shape of data.
This lesson shows the full pipeline from data to topological invariants, then steps back to survey the grand arc of algebraic topology from Euler to the present.
Watch the complete pipeline: start with a point cloud, grow balls around each point (controlled by ε), build the Vietoris-Rips complex, and track which topological features (components, loops) persist across scales. Long bars in the barcode represent genuine features; short bars are noise.
Key insight: Persistent homology turns a continuous parameter (ε) into a discrete algebraic summary (the barcode). Features that persist over a wide range of ε are topologically significant — this is the stability theorem in action.
Pick an image and drag the threshold slider through the gray levels. Watch holes appear and vanish: the letter “A” has one, “B” has two.
The letter A has one hole — the triangular space inside
Apply sublevel set filtration to images. As the threshold sweeps through gray levels, topological features appear and disappear. The letter “A” has one hole (H&sub1; = 1); “B” has two (H&sub1; = 2). Topology detects these features regardless of font, size, or distortion.
Key insight: Topology provides features that are invariant under continuous deformation. This makes topological descriptors robust to noise, rotation, and scaling — unlike pixel-based features.
Click any milestone to read its story, from Euler's bridges to Perelman's proof.
From Euler's bridges to Perelman's proof, algebraic topology has grown from a geometric curiosity into one of mathematics' most powerful frameworks.
Key insight: Emmy Noether's 1925 insight — that homology should be understood as groups, not just numbers — transformed the entire field. Every concept in this module flows from that algebraic perspective.
Spot a mistake or have an idea?