semanticsimilaritytable (Transact-SQL)

Стаття
02/28/2023

Applies to: SQL Server

Returns a table of zero, one, or more rows for documents whose content in the specified columns is semantically similar to a specified document.

This rowset function can be referenced in the FROM clause of a SELECT statement like a regular table name.

Transact-SQL syntax conventions

Syntax

SEMANTICSIMILARITYTABLE  
    (  
    table,  
    { column | (column_list) | * },  
    source_key  
    )

Arguments

table
Is the name of a table that has full-text and semantic indexing enabled.

This name can be a one to four part name, but a remote server name is not allowed.

column
Name of the indexed column for which results should be returned. Column must have semantic indexing enabled.

column_list
Indicates several columns, separated by a comma and enclosed in parentheses. All columns must have semantic indexing enabled.

*
Indicates that all columns that have semantic indexing enabled are included.

source_key
Unique key for the row, to request results for a specific row.

The key is implicitly converted to the type of the full-text unique key in the source table whenever possible. The key can be specified as a constant or a variable, but cannot be an expression or the result of a scalar sub-query.

Table Returned

The following table describes the information about similar or related documents that this rowset function returns.

Matched documents are returned on per-column basis if results are requested from more than one column.

Column_name	Type	Description
source_column_id	int	ID of the column from which a source document was used to find similar documents. See the COL_NAME and COLUMNPROPERTY functions for details on how to retrieve column name from column_id and vice versa.
matched_column_id	int	ID of the column from which a similar document was found. See the COL_NAME and COLUMNPROPERTY functions for details on how to retrieve column name from column_id and vice versa.
matched_document_key	* This key matches the type of the unique key in the source table.	Full-text and semantic extraction unique key value of the document or row that was found to be similar to the specified document in the query.
score	REAL	A relative value for similarity for this document in its relationship to all the other similar documents. The value is a fractional decimal value in the range of [0.0, 1.0] where a higher score represents a closer match and 1.0 is a perfect score.

General Remarks

For more information, see Find Similar and Related Documents with Semantic Search.

Limitations and Restrictions

You cannot query across columns for similar documents. The SEMANTICSIMILARITYTABLE function only retrieves similar documents from the same column as the source column, which is identified by the source_key argument.

Metadata

For information and status about semantic similarity extraction and population, query the following dynamic management views:

Security

Permissions

Requires SELECT permissions on the base table on which the full-text and semantic indexes were created.

Examples

The following example retrieves the top 10 candidates who are similar to a specified candidate from the HumanResources.JobCandidate table in the AdventureWorks2022 sample database.

SELECT TOP(10) KEY_TBL.matched_document_key AS Candidate_ID  
FROM SEMANTICSIMILARITYTABLE  
    (  
    HumanResources.JobCandidate,  
    Resume,  
    @CandidateID  
    ) AS KEY_TBL  
ORDER BY KEY_TBL.score DESC;

Поділитися через