Term by Document Matrix, how do I determine which document corresponds with document number

William29 — Mon, 03 Jul 2023 06:57:54 GMT

I have a dataset with a text variable and I ran the 'Text Parsing and Document Discovery" feature under SAS VIYA text analytics and I produced term by document matrix as below:

How do I interpret the document number? Is there a simple way of determining which document corresponds to a particular document number? Document number does not appear to be row number so I am wondering if I can get some guidance with how I associate a particular document number with a particular document/row number

Re: Term by Document Matrix, how do I determine which document corresponds with document number

sbxkoenk — Tue, 04 Jul 2023 15:00:14 GMT

Hello,

I would never do this analysis without including a Document ID (to be able to find everything back afterwards and to be able to link / map everything).

I wasn't even aware you could do this analysis without including a Document ID.

The procedure that is run in the background is PROC TEXTMINE.

PROC TEXTMINE has a DOC_ID Statement. I think that DOC_ID Statement is mandatory.

DOC_ID Statement

DOC_ID variable;

The DOC_ID statement specifies the variable that contains the ID of each document. In the input data table, each row corresponds to one document. The ID of each document must be unique; it can be either a number or a string of characters.

Kind regards,
Koen

topic Re: Term by Document Matrix, how do I determine which document corresponds with document number in SAS Data Science

Term by Document Matrix, how do I determine which document corresponds with document number

Re: Term by Document Matrix, how do I determine which document corresponds with document number

DOC_ID Statement