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. ID of the column from which a source document was used to find similar documents. Matched documents are returned on per-column basis if results are requested from more than one column. The following table describes the information about similar or related documents that this rowset function returns. The key can be specified as a constant or a variable, but cannot be an expression or the result of a scalar sub-query. The key is implicitly converted to the type of the full-text unique key in the source table whenever possible. Unique key for the row, to request results for a specific row. Indicates that all columns that have semantic indexing enabled are included. All columns must have semantic indexing enabled. Indicates several columns, separated by a comma and enclosed in parentheses. Column must have semantic indexing enabled. Name of the indexed column for which results should be returned. This name can be a one to four part name, but a remote server name is not allowed. Is the name of a table that has full-text and semantic indexing enabled. Transact-SQL syntax conventions Syntax SEMANTICSIMILARITYTABLE This rowset function can be referenced in the FROM clause of a SELECT statement like a regular table name. Returns a table of zero, one, or more rows for documents whose content in the specified columns is semantically similar to a specified document.
0 Comments
Leave a Reply. |