Homology is a systematic method in mathematics for converting geometric or algebraic objects into sequences of algebraic invariants. Concretely, it produces a sequence of groups (or modules) that reflect global structural features of the original object. These groups are usually abelian; typical targets are abelian groups or modules over a chosen coefficient ring. Homology is widely used to detect and classify different kinds of "holes" in spaces, to compute numerical invariants, and to relate topology with algebra.

Basic idea and formal definition

The construction begins with a chain complex: a collection of abelian groups C_n together with boundary maps d_n: C_n → C_{n-1} satisfying d_{n} ∘ d_{n+1} = 0. The nth homology group is defined as the quotient H_n = ker(d_n) / im(d_{n+1}). Intuitively, cycles (elements of ker d_n) represent closed n-dimensional features, while boundaries (elements of im d_{n+1}) are those cycles that bound something of one higher dimension and so do not represent genuine holes.

Common variants and constructions

  • Simplicial homology: defined for spaces built from simplices (triangles, tetrahedra, etc.).
  • Singular homology: defined for arbitrary topological spaces using continuous maps from standard simplices; it is homotopy invariant and widely applicable.
  • Cellular homology: adapted to CW complexes and often simplifies computations by using the cell structure.
  • Relative and reduced homology: variants that compare a space with a subspace or adjust the zeroth group for connectedness information.

Homology groups are functorial: a continuous map between spaces induces group homomorphisms between their homology groups. This functoriality leads to many computational and theoretical tools such as exact sequences, the Mayer–Vietoris sequence, the universal coefficient theorem, and the Künneth formula for product spaces.

Historical context and axiomatic approach

The idea of assigning algebraic objects to topological spaces to capture global properties goes back to work of Henri Poincaré and others at the turn of the 20th century. In the mid-20th century, algebraic topologists formalized and expanded these ideas; axioms for homology theories (the Eilenberg–Steenrod axioms) clarified what properties a reasonable theory should satisfy and guided the development of generalized homology theories.

Examples and significance

Simple computations illustrate the meaning of homology: the circle S^1 has H_0 ≅ Z and H_1 ≅ Z (one connected component and one independent 1-dimensional hole), while higher groups vanish; an n-sphere S^n has H_0 ≅ Z and H_n ≅ Z with all other H_k = 0. More complicated spaces like tori or knot complements have homology groups that record multiple independent cycles and possible torsion elements. The ranks of free parts of homology groups are the Betti numbers, which appear in formulas for the Euler characteristic and in quantitative descriptions of topology.

Relations and distinctions

Homology is easier to compute in many cases than homotopy groups, because homology groups are always abelian and admit linear-algebra-style techniques. However, homology can miss certain fine structures—different spaces can have the same homology but different homotopy type. For manifolds, Poincaré duality relates homology and cohomology, bridging geometric orientation and algebraic invariants. There are also algebraic variants: topological space homology, group homology, and generalized theories used in cohomology and K-theory.

Further study typically covers computational tools (cellular chains, Mayer–Vietoris), the behavior with different coefficients, and generalized homology theories that extend the classical axioms. For introductions and technical references see standard texts and surveys in algebraic topology or follow introductory resources on topological methods.

Overall, homology provides a bridge from geometry to algebra: a compact, computable way to quantify and compare the "holes" and global features of spaces and other mathematical structures.