Canonicalization

Canonicalization is the process of converting data into a standard or canonical form. The goal is to remove variation that is not essential to the meaning or function of the data, so that two representations that convey the same information can be treated as equivalent.

In computing and text processing, canonicalization includes Unicode normalization (such as NFC, NFD, NFKC, NFKD) to ensure that strings with the same meaning are identical at the code point level. It also covers case folding, diacritic stripping, whitespace normalization, and locale-sensitive rules.

In information retrieval and web technologies, canonicalization helps prevent duplicate content from diluting search relevance. Canonical URLs specify the preferred address for a resource; search engines use rel="canonical" to consolidate signals, while web crawlers avoid indexing duplicates.

In data management, canonicalization maps diverse data into a canonical data model or schema. This supports deduplication, data integration, and consistent analytics by transforming inputs to standardized representations before storage or processing.

In security, improper canonicalization can create vulnerabilities such as path traversal or input bypass, where different representations are treated differently by subsystems. Safe design requires consistent, fully specified normalization across components.

In mathematics and computer algebra, a canonical form is a standard representation that uniquely identifies an object up to isomorphism or equivalence. Examples include Jordan canonical form for matrices, canonical forms of polynomials, or normal forms in logic.

See also: normalization, data normalization, Unicode normalization, URL normalization, deduplication.