tsrank

tsrank is a term that appears in several technical contexts and does not refer to a single, universally defined concept. Broadly, it denotes a ranking score or function associated with data that is used to order items by relevance, similarity, or importance. Because different domains package or expose the concept under similar names, the precise behavior of tsrank is context-dependent.

In PostgreSQL, ts_rank is a built-in function used in full-text search. Given a tsvector that encodes document text and a tsquery representing the search terms, ts_rank computes a floating-point score indicating how well the document matches the query. The score can reflect term frequency and positional information, and several normalization options or weighting schemes may alter the result. A related variant, ts_rank_cd, applies cover-density normalization to emphasize dense matches.

In other domains, tsrank may appear as a generic term for a time-series or text-scoring ranking method or as the name of a library function in a specific software package. In such cases, tsrank typically yields a numeric score that is then used to sort items, select the top results, or feed into downstream modeling tasks. The exact formula, normalization, and adopted features vary by implementation.

See also ts_rank and ts_rank_cd in PostgreSQL, full-text search ranking, ranking algorithms, and time-series analysis.