如何使用 R 检查字符串是否是维基百科文章的标题？

Question

假设我有一个字符串列表：

strings <- c("dog", "cat", "animal", "bird", "birds", "bqpohd", "ohphha", "mqphihpha", "aphhphohpa", "pohha")

我想检查一下这些字符串是否是维基百科文章的标题。

这是一个解决方案，但我认为这不是完成长列表任务的最快方法：

results.df <- CheckIfAStringIsTheTitleOfAWikipediaArticle(strings)
View(results.df)
CheckIfAStringIsTheTitleOfAWikipediaArticle <- function(strings){
  start_time <- Sys.time()
    Check <- function(string){
    GetPageID <- function(string){
  query <- paste0("https://en.wikipedia.org/w/api.php?", 
                  "action=query", "&format=xml", "&titles=", string)
  answer <- httr::GET(query)
  library(xml2)
  library(httr)
  page.xml <- xml2::read_xml(answer)
  nodes <- xml_find_all(page.xml, ".//query//pages//page")
  pageid <- xml_attr(nodes, "pageid", ns = character(), 
                     default = NA_character_)
  return(pageid)
  }

  IsValidPageName <- function(string){
    pageid<- GetPageID(string)
    if(!is.na(pageid)){return(TRUE)}
    else{return(FALSE)}
  }

  boolean <- IsValidPageName(string)
  return(boolean)
  }

  validTitle <- unlist(lapply(strings, Check))
  results.df <- data.frame(strings, validTitle)
  end_time <- Sys.time()
  time <- end_time - start_time
  print(time)
  return(results.df)
}

非常感谢您的帮助！

Answer 1

这是基本的 R 方法。

将所有 English titles from Wikipedia 下载到临时文件中。然后将它们扫描到内存中。大约 1.2 Gb。

我假设您不关心大小写，因此我们需要使用 tolower 将标题更改为全部小写。然后只需使用 %in%.

strings <- c("dog", "cat", "animal", "bird", "birds", "bqpohd", "ohphha", "mqphihpha", "aphhphohpa", "pohha")

url <- "http://dumps.wikimedia.org/enwiki/latest/enwiki-latest-all-titles-in-ns0.gz"
tmp <- tempfile()
download.file(url,tmp)
titles <- scan(gzfile(tmp),character())
titles <- tolower(titles)
strings[strings %in% titles]
[1] "dog"    "cat"    "animal" "bird"   "birds" 

#Reasonably fast
system.time(strings[strings %in% titles])
   user  system elapsed 
  1.494   0.029   1.525

如何使用 R 检查字符串是否是维基百科文章的标题？

How to check if a string is the title of a Wikipedia article with R?

r

string

wikipedia-api

page-title