R中不区分大小写拼写检查咨询：hunspell包优化及替代方案

Nice question! Let's break this down—first how to keep using hunspell while ignoring case, then some alternative packages that handle this more seamlessly.

1. 继续用hunspell：手动统一大小写来绕过限制 #

You're right that hunspell_check() is case-sensitive by default, but the fix is straightforward: since spelling correctness doesn't depend on case, just convert all your target words to lowercase (or uppercase) before running the check. You can then map the results back to your original words if needed.

Example code:

library(hunspell)

# Mixed-case words to check
words <- c("Apple", "APPLE", "appel", "Banana", "BANANA", "banan")

# Convert to lowercase and run check
case_insensitive_check <- hunspell_check(tolower(words))

# Label results with original words for clarity
names(case_insensitive_check) <- words
case_insensitive_check

This will return TRUE for correctly spelled words regardless of their original case, and FALSE for misspellings.

2. Alternative R packages for case-insensitive spelling checks #

If you'd rather avoid manual case conversion, these packages have built-in support or more intuitive APIs:

• spelling package (built on hunspell):
  This package wraps hunspell with cleaner functions. You can use spell_check_text() with ignore.case = TRUE to skip case differences automatically:

  library(spelling)
  spell_check_text(words, ignore.case = TRUE)
  

  It returns a neat data frame with misspelled words and suggested corrections.

• textclean package:
  Designed for text preprocessing, its check_spelling() function includes an explicit ignore.case parameter:

  library(textclean)
  check_spelling(words, ignore.case = TRUE)
  

  This function highlights misspellings and offers fixes, making it great for batch text cleaning.

• qdapTools package:
  If you want to directly get corrected words instead of just checks, use the correct() function with case insensitivity:

  library(qdapTools)
  corrected_words <- correct(words, ignore.case = TRUE)
  corrected_words
  

  It replaces misspelled words with the most likely correct version, ignoring case differences.

内容的提问来源于stack exchange，提问作者Harshad Patil