topic Re: How to get nouns from a given string in SAS Programming

How to get nouns from a given string

Azeem112 — Mon, 17 Oct 2022 08:22:42 GMT

I have some string data and I want to extract the nouns from those strings.

data text;
infile datalines delimiter=",";
length Id $5 string $100;
input Id String $;
datalines;
10, Ahmad is a student in fifth grade
11, His mother is a nurse in a hospital
12, His aunt is a kindergarten teacher
;
run;

Now I want to extract the nouns from this data.

I want a table with the below data as

ID Noun

10 student grade

11 mother nurse

12 aunt kindergarten teacher

Is there any NLP library or function in SAS that can do this?

Re: How to get nouns from a given string

andreas_lds — Mon, 17 Oct 2022 09:49:56 GMT

I don't know any function that can do this, would be interesting to maintain such a function, which, of course, needs to support multiple languages.

Maybe such a function exists in a special sas-component.

Maybe you can find a list of nouns in the www.

If you have such a list, the program could have the following steps:

read the list
create an informat (see code below)
check each word

data nouns_fmt;
   set list(rename=(noun = start)) end=jobDone;
   retain FmtName 'IsNoun' Type 'i' Hlo 'U' Label 1;
   
   Start = upcase(start);
   output;
   
   if jobDone then do;
      HLO = 'O';
      Label = 0;
      output;
   end;
run;

proc format cntlin= nouns_fmt;
run;

Re: How to get nouns from a given string

ballardw — Mon, 17 Oct 2022 14:49:11 GMT

Ahmed is a Proper Noun. Is there are RULE that proper nouns are not to be included?

What is the RULE that Hospital is not considered as a noun in the second observation?

What would the result be for a sentence like:

I grade papers.

Natural language processing is a pain. English may be the worst language to attempt this with because so many "nouns" are also adjectives, adverbs or verbs.

There are also problems related with separating proper nouns from nouns with the fads for "yewneek" spellings of children's names where more names are place names, drinks, occupations and other non-traditional words. Remembering of course than many current proper names are nouns or job descriptions in older languages or versions of English.

Re: How to get nouns from a given string

Azeem112 — Tue, 18 Oct 2022 06:40:06 GMT

There is not any rule regarding nouns and proper nouns. I just need to get noun information for analytics. it could be both nouns and proper nouns. I extracted the nouns from an online website.

Re: How to get nouns from a given string

Kurt_Bremser — Tue, 18 Oct 2022 06:57:04 GMT

So you "just" need to look for words you already have in a dataset?

Re: How to get nouns from a given string

Azeem112 — Tue, 18 Oct 2022 07:16:41 GMT

I just need to look fr words that user enter in a string and I need to extract the nouns from it.

Re: How to get nouns from a given string

Patrick — Tue, 18 Oct 2022 07:18:07 GMT

It looks like SAS Visual Text Analytics provides functionality in this area.

Re: How to get nouns from a given string

Kurt_Bremser — Tue, 18 Oct 2022 08:59:04 GMT

Quote from you:

I extracted the nouns from an online website

Now, what is it? Manually entered, or retrieved from a website?

Please describe in detail(!!!) your whole process.

Re: How to get nouns from a given string

Azeem112 — Wed, 19 Oct 2022 09:57:47 GMT

I get that manually from a website.

Re: How to get nouns from a given string

Azeem112 — Wed, 19 Oct 2022 09:59:47 GMT

I searched "Get nouns from a string online" and the first website does this for me

Re: How to get nouns from a given string

Azeem112 — Tue, 01 Nov 2022 13:13:28 GMT

I wrote a python code that returns noun words using nltk lbirary and passed that string to it as parameter. I executed that code using pipe command and saved the output in a sas dataset.