topic Return the word after and before a keyword in a column in SAS Data Science

Return the word after and before a keyword in a column

shivamnegi92 — Mon, 29 Jan 2018 19:00:17 GMT

I'm looking to create a column that contains the word 3 words before and 3 words directly after a keyword in my target column "Apple Description"

For example, if the keyword I would like to search for is "Apple" then

The market has apples and alot of bananas ==> column 1 ->> "The market has" . column 2 ->> " and alot of "

I love to eat a lot of apples ==> "a lot of"

I like orange juice ==> ""

And I would like to apply that to multiple keywords like "apple" or "apple like" or "apple fruit" altogether.

Thanks!

data want;
  length before after $200;
  string="a b c d e trigger f g h i j";
  do i=1 to countw(string," ");
    if scan(string,i," ")="trigger" then do;
      before=scan(string,i-5," ");
      after=scan(string,i+5," ");
      before2=scan(string,i-4," ");
      after2=scan(string,i+4," ");
    end;
  end;
run;

Re: Return the word after and before a keyword in a column

novinosrin — Mon, 29 Jan 2018 19:27:14 GMT

/*fun stuff*/

data want;
string='The market has apples and alot of bananas';
length before after $50;
  do i=1 to countw(string," ");
      if  scan(string,i," ")='apples' then do;
	  do n=i-3 to i-1 ;
	  before=catx(' ',before,scan(string,n));
      end;
	
	  do n=i+1 to i+3;
	  after= catx(' ',after,scan(string,n));
		end;
		end;
	end;
drop n i;
run;

Re: Return the word after and before a keyword in a column

shivamnegi92 — Mon, 29 Jan 2018 19:45:31 GMT

That is very helping. Thanks , that partially solved my problem. Suppose, instead of a string, I have to apply it to a column, how we can go ahead with this code.

By that, I mean, suppose instead of string, we have a dataframe with a specific column we are targetting , suppose -- "Apple summary". Then if you wanna find out 3 words before and after a keyword -"Apple", And create 2 columns with --3 WORDS BEFORE "apple" and 3 words after apple, then how you will go ahead.

Sorry, I have limited understanding about SAS.

Thanks again for quick reply.