How to keep specific group of words or phrases of a text column in R?

Question

I have a dataframe with a text column and I would like to create another column only with specific words or phrases matching the text column. Let's say I have these 4 rows in the dataframe:

   TEXT_COLUMN
1 "discovering the hidden themes in the collection."
2 "classifying the documents into the discovered themes."
3 "using the classification to organize/summarize/search the documents."
4 "alternatively, we can set a threshold on the score"

And, on the other hand, I have a list of words and phrases I want to keep. For example:

x <- c("hidden themes", "the documents", "discovered themes", "classification to organize", "search")

So, I would like to create a new column "KEYWORDS" with the words in "x" which match the text column separated by a comma:

   TEXT_COLUMN                                                             |  KEYWORDS
1 "discovering the hidden themes in the collection."                       |  "hidden themes"
2 "classifying the documents into the discovered themes."                  |  "the documents", "discovered themes"
3 "using the classification to organize/summarize/search the documents."   |  "classification to organize", "search"
4 "alternatively, we can set a threshold on the score"                     |  NA

Do you know any way to do this?

Thank you very much in advance.

Answer 1

An option is to create a pattern from 'x' by joining with str_c

library(stringr)
library(dplyr)
pat <- str_c("\\b(", str_c(x, collapse="|"), ")\\b")

Then, using this pattern, extract the substring from the 'TEXT_COLUMN' into a list column of vectors

df1 <- df1 %>% 
      mutate(KEYWORDS = str_extract_all(TEXT_COLUMN, pat))

-output

df1
#TEXT_COLUMN                                          KEYWORDS
#1                     discovering the hidden themes in the collection.                                     hidden themes
#2                classifying the documents into the discovered themes.                  the documents, discovered themes
#3 using the classification to organize/summarize/search the documents. classification to organize, search, the documents
#4                   alternatively, we can set a threshold on the score                                                  

data

df1 <- structure(list(TEXT_COLUMN = c("discovering the hidden themes in the collection.", 
"classifying the documents into the discovered themes.", "using the classification to organize/summarize/search the documents.", 
"alternatively, we can set a threshold on the score")), 
class = "data.frame", row.names = c("1", 
"2", "3", "4"))