Geometry · Concept hub

Homology

1895 CE19th-century France (Poincaré)

Concept

A way to record holes in a space with algebra. H₀ detects connected components, H₁ detects loops that do not bound a filled region, and H₂ detects higher-dimensional enclosed structure. Rather than merely counting holes, homology uses groups and boundary maps to distinguish topology that survives stretching and bending.

Understand it in one breath

"Stretch all you like with rubber — the number of holes does not change." A donut (genus 1) has H₁ = ℤ²; a sphere has H₁ = 0. A closed loop that is not the boundary of a filled-in region records a one-dimensional hole. In the 21st century, persistent homology turns this idea into tools for analyzing the shape of data.

At a glance

abH₁(T²) = ℤ ⊕ ℤ — two independent cyclesNeither a nor b bounds a smaller region → genuine holes

Key formula

Hn(X)=ker(n)/im(n+1)H_n(X) = \ker(\partial_n) / \mathrm{im}(\partial_{n+1})

Modern applications

Protein-folding analysis, topological data analysis, and coverage guarantees in sensor networks.

Beyond MathVoyage

Loading…