RECOMMEND Procedure

ADDTABLE Statement

The ADDTABLE statement specifies a table for a recommender system that the SAS LASR Analytic Server uses to derive internal tables. The ADDTABLE statement is used most often to identify a table that contains user-item-ratings information.

Syntax

Required Argument

ADDTABLE Statement Options

Syntax

ADDTABLE libref.member-name </ option(s)>;

Required Argument

table

specifies a source table that the SAS LASR Analytic Server uses to generate internal tables. Specify the table name in Lasrlib.Name format, where the Lasrlib value is the libref of the SAS LASR Analytic Server on which the recommender system is defined.

Requirement

The table that you specify must already be in memory.

ADDTABLE Statement Options

You can specify the following optional arguments after the slash (/) in the ADDTABLE statement. These options apply to the recommender system that is used by the ADDTABLE statement.

SYSTEM=recommender-system

specifies the name of the recommender system in the SAS LASR Analytic Server that the procedure works with. Specify a two-level name, similar to a LIBNAME.MEMBER construct.

To work with an existing recommender system, the name that you specify identifies the application in the SAS LASR Analytic Server. When you create a new recommender system, the specified value becomes the name of that new system.

Alias	RECOM=
Default	RECOM.SYSTEM

TYPE=ITEM

TYPE=USER

TYPE=RATING

specifies what type of information the table contains. A table of TYPE=ITEM contains the item column that was specified in the ADD statement and content information for the items. A table of TYPE=USER contains the user column that was specified in the ADD statement and content information for the users. A table of TYPE=RATING contains user-item-ratings information.

Default

RATING

VARS=(variable-list)

specifies a list of one or more variables to transfer to the internal tables that the SAS LASR Analytic Server derives from the in-memory table. For example, if you are adding a table of item content, then only a subset of the variables are useful to the recommender system. List the useful variables in the VARS= option.

Default

Transfer all variables