Question

继续我的文本挖掘分析我得到了错误即我尝试干净的语料库

my.stopwords=read.table("path to my stopword list") stops=as.character(my.stopwords$V1)
generateCorpus <- function(df, stops = c()) {
  #corpus <- Corpus(VectorSource(df))

  myCorpus <- Corpus(VectorSource(df))
  if (Sys.info()['sysname']=="Darwin"){
    -        myCorpus <- tm_map(myCorpus, function(x) iconv(x, to='UTF-8-MAC', sub='byte'))   
    +        myCorpus <- tm_map(myCorpus, content_transformer(function(x) iconv(x, to='UTF-8-MAC', sub='byte')))   
  }
  if (Sys.info()['sysname']=="Windows"){
    -        myCorpus <- tm_map(myCorpus, function(x) iconv(enc2utf8(x), sub = "byte"))  
    +        myCorpus <- tm_map(myCorpus, content_transformer(function(x) iconv(enc2utf8(x), sub = "byte")))
  }

  # convert to lower case
  cat("Converting to lowercase... ")
  -    myCorpus <- tm_map(myCorpus, tolower)
  +    myCorpus <- tm_map(myCorpus, content_transformer(tolower))
  cat("done!\n")
  # remove numbers
  cat("Removing digits and URLs... ")
  -    myCorpus <- tm_map(myCorpus, removeNumbers)
  +    myCorpus <- tm_map(myCorpus, content_transformer(removeNumbers))
  # remove URLS
  removeURL <- function(x) gsub('"(http.*) |(http.*)$|\n', "", x)
  cat("done!\n")
  -    myCorpus <- tm_map(myCorpus, removeURL) 
  +    myCorpus <- tm_map(myCorpus, content_transformer(removeURL))

}
my.corpus <- generateCorpus(myCorpus, stops)

结果我收到了这条消息

enc2utf8（x）出错：参数不是字符向量

这是什么意思？

更新

> if (Sys.info()['sysname']=="Windows"){
+            myCorpus <- tm_map( myCorpus, function(x) iconv(enc2utf8(as.character(x)), sub = "byte")
+     )
+     myCorpus <- tm_map(myCorpus, content_transformer(function(x) iconv(enc2utf8(as.character(x)), sub = "byte")))
+   }
 Show Traceback

 Rerun with Debug
 Error in UseMethod("content", x) : 
  no applicable method for 'content' applied to an object of class "character"

我认为内容中的主要问题

Error in UseMethod("content", x) : 
  no applicable method for 'content' applied to an object of class "character"
Called from: content(x)
Browse[1]> 

function (x) 
UseMethod("content", x)

谁决定了这个问题？

Answer 1

似乎tm_map不会迭代字符，而是文档。

可以看出这一点

data(crude)
tm_map(crude, function(x) browser())

然后，检查x，

x
#<<PlainTextDocument>>
#...
enc2utf8(x)
#error in enc2utf8(x) : argumemt is not a character vector

首先应用as.character就好了，也就是说

myCorpus <- tm_map(
    myCorpus, 
    function(x) {
        content(x) <- iconv(enc2utf8(as.character(content(x))), sub = "byte")
        return(x)
    }
)

可能会成功。

Answer 2

我在这个网站找到了解决方案 http://www.listendata.com/2014/11/create-wordcloud-with-r.html

我改变了任何字符串，之后，我得到了所需的结果

removeURL <- function(x) gsub('"(http.*) |(http.*)$|\n', "", x)
myCorpus <- tm_map(myCorpus, removeURL)

kb.tf <- list(weighting = weightTf, stopwords  = stopwords,
              removePunctuation = TRUE,
              tolower = TRUE,
              minWordLength = 4,
              removeNumbers = TRUE, stripWhitespace = TRUE,
              stemDocument= TRUE)

myCorpus <- tm_map(myCorpus, PlainTextDocument) 

generateCorpus <- function(df, kb.tf = c()) {
myCorpus <- Corpus(VectorSource(df))
}
my.corpus <- generateCorpus(myCorpus, kb.tf)

我想我确实做出了贡献，因为我看到类比帖子，这个问题很热

错误：参数不是字符向量

2 个答案: