从 R 中的文本中删除单词和符号

Question

Consider the following example. Is that possible to delete stopwords from text ?

library(tm)
text <- c("this is exercise for text mining ≤µm ≥°f ±μgm")
stopwords <- ("≤µm", "≥°f", "±μgm")

Answer 1

你可以像下面这样尝试gsub

gsub(paste0(stopwords, collapse = "|"),"",text)

Answer 2

First, you have some errors in your example strings. text is missing quotation marks and stopwords is missing the c before the brackets.

text <- c("this is exercise for text mining ≤µm ≥°f ±μgm")
stopwords <- c("≤µm", "≥°f", "±μgm")

You can remove the values in stopwords from your string using stringr as below:

library(stringr)
str_replace_all(text, paste(stopwords, collapse = "|"), "")

Answer 3

Since you are going to be doing text mining, you may be wanting to convert your input string into a vector of words. If so, you can easily remove the stopwords by subsetting.

library(stringr)
text <- c("this is exercise for text mining ≤µm ≥°f ±μgm")
stopwords <- c("≤µm", "≥°f", "±μgm")
text <- unlist(str_split(text, " "))
text[!(sapply(text, function (x) any(str_detect(stopwords, x))))]

If your work has you putting your words in a data.frame or similar, then there is another way:

library(dplyr)
library(stringr)
text <- c("this is exercise for text mining ≤µm ≥°f ±μgm")
stopwords <- c("≤µm", "≥°f", "±μgm")
text <- unlist(str_split(text, " "))
data.frame(words = text) %>% anti_join(data.frame(words = stopwords))