CASE /WHEN is SQL syntax, not DATA step.     Even at that the "new column" comes after the END of the Case construct Case when <condition> result end as NEWCOLUMN   Data step would use IF/Then or SELECT If variable = value then newvariable=something; ELSE if variable= othervalue then Newvariable=somethingelse; <else ...> Or Select (variable); when (value) NewVariable=something; when (othervalue) Newvariable=somethingelse; otherwise <what to do when none of the listed values are encountered>; end; ... View more

Thnaks Russ and all the friends that contributed to my question.    You were right, i checked carefully the code that is downloaded from the SAS Visual Text Analytics and i can score a new data set by filling in the macro variables. Actually the scored data set is saved in a cas library called sascas1 that is created by the code, although i set the library to casuser. This is why i could not find the scored data set and i initiated the post. So if someone else tries this action the scored data set can be found in the sascas1 cas library.   /* specifies the CAS library to write the score output table. You must modify the value to provide the name of the library that will contain the output table that the score code produces. */ %let output_caslib_name = "casuser";   cas sascas1 host=&cas_server_hostname port=&cas_server_port; libname sascas1 cas sessref=sascas1 datalimit=all;     Thanks! ... View more

Hello @azaras  “pos” in the code is just the name of the table that gets created to contain the position of child terms in each document of the document collection by the “offset” parameter. It records the start and end position for a term in a document.   The count of child terms in your pos table is higher than in the terms table since a term can appear in multiple documents. The terms table lists the individual terms once and shows the frequency of terms in the document collection. The difference in count is not due to dropped terms. The pos table has a column “_Document_” that you can check if you want to verify what is occurring.   As for the stop list The Environment Manager > Data > ReferenceData folder has the default stoplists. The en_stoplist file for example contains the default stop words for English. There are over 1200 words in that list. Customers often create their own stop word lists for domain specific document collections. ... View more