Search for and extract words into new column - r

Question

I have a dataset with 63,000 rows in R. One of the columns contained a list of words in the format '"Fireplace", "Garage", "One story with balcony", "Off street parking",' etc. They are property characteristics listed from sale websites.
I want to extract words from this column and create a new column that has a '0' or a '1' if the word is present or not (creating a dummy variable for regression). Once that has been completed, I want to be able to merge some of those columns together (ie take 'parking' 'Parking' 'garage' 'Garage' columns and merge them into one that includes all parking and garages). I'm assuming that R is sensitive to upper and lowercase characters but even if not, I need to be able to merge 'parking' and 'garage' together, for example.
This is for a hedonic pricing method so I need as many property characteristic variables as possible.
I don't know how to create new dummy variables or merge them into one column once I have, so am struggling. Would appreciate any help.

Answer 1

Is this what you are looking for?
library(tidyverse)
data.frame(txt) %>%
# tidy up `txt`:
mutate(txt = gsub("(?![, ])\\W", "", txt, perl = TRUE)) %>%
# split into rows
separate_rows(txt, sep = ",") %>%
# extract keywords matched:
mutate(keywords = str_extract(txt, "(?i)Parking|Garage|Garden|Freehold|Fireplace|Balcony"))
# A tibble: 19 × 2
txt keywords
<chr> <chr>
1 "Stunning seaside location" NA
2 " 24hour emergency call system and secure video entry" NA
3 " Mature landscaped gardens with large terraces and seating areas" garden
4 " Walk out balconies to selected apartments" NA
5 " Beautifully decorated homeowners8099 lounge" NA
6 " Parking spaces and car ports are available via an annual permit" Parking
7 " Wheelchair access" NA
8 " Lifts to all floors" NA
9 " Fire detection" NA
10 " Intruder alarm" NA
11 " Village Location" NA
12 " 4 Bedrooms" NA
13 " Gardens" Garden
14 " Balcony" Balcony
15 " On streetresidents parking" parking
16 " Central heating" NA
17 " Double glazing" NA
18 " Fireplace" Fireplace
19 " Ruralsecluded" NA
Data:
txt <- '"[\"Stunning seaside location\", \"24-hour emergency call system and secure video entry\", \"Mature landscaped gardens with large terraces and seating areas\", \"Walk out balconies to selected apartments\", \"Beautifully decorated homeownersâ\200\231 lounge\", \"Parking spaces and car ports are available via an annual permit\", \"Wheelchair access\", \"Lifts to all floors\", \"Fire detection\", \"Intruder alarm\"]", "[\"Village Location, 4 Bedrooms, Garden(s)\"]", "[\"Balcony\", \"On street/residents parking\", \"Central heating\", \"Double glazing\", \"Fireplace\", \"Rural/secluded\"]"'
If there may be more than 1 keyword per substring, then use str_extract_all in this way:
data.frame(txt) %>%
mutate(txt = gsub("(?![, ])\\W", "", txt, perl = TRUE)) %>%
separate_rows(txt, sep = ",") %>%
mutate(keywords = str_extract_all(txt, "(?i)Parking|Garage|Garden|Freehold|Fireplace|Balcony")) %>%
unnest(where(is.list), keep_empty = TRUE)
EDIT:
If the OP is looking to obtain a variable for each keyword, then this works:
data.frame(txt) %>%
mutate(txt = gsub("(?![, /])\\W", "", txt, perl = TRUE)) %>%
separate_rows(txt, sep = ", ") %>%
mutate(keywords = str_extract_all(txt, "(?i)Parking|Garage|Garden|Freehold|Fireplace|Balcony")) %>%
# unnest listed items:
unnest(where(is.list), keep_empty = TRUE) %>%
# capitalize initial letter:
mutate(keywords = sub("^(.)", "\\U\\1", keywords, perl = TRUE)) %>%
# cast each keaword into its own column:
pivot_wider(names_from = keywords, values_from = keywords,
values_fn = function(x) 1, values_fill = 0)
# A tibble: 19 × 6
txt `NA` Garden Parking Balcony Fireplace
<chr> <dbl> <dbl> <dbl> <dbl> <dbl>
1 Stunning seaside location 1 0 0 0 0
2 24hour emergency call system and secure video entry 1 0 0 0 0
3 Mature landscaped gardens with large terraces and seating areas 0 1 0 0 0
4 Walk out balconies to selected apartments 1 0 0 0 0
5 Beautifully decorated homeowners8099 lounge 1 0 0 0 0
6 Parking spaces and car ports are available via an annual permit 0 0 1 0 0
7 Wheelchair access 1 0 0 0 0
8 Lifts to all floors 1 0 0 0 0
9 Fire detection 1 0 0 0 0
10 Intruder alarm 1 0 0 0 0
11 Village Location 1 0 0 0 0
12 4 Bedrooms 1 0 0 0 0
13 Gardens 0 1 0 0 0
14 Balcony 0 0 0 1 0
15 On street/residents parking 0 0 1 0 0
16 Central heating 1 0 0 0 0
17 Double glazing 1 0 0 0 0
18 Fireplace 0 0 0 0 1
19 Rural/secluded 1 0 0 0 0