Question

我正在尝试使用tm包在R上开发基本文本分析。

输入文件：csv文件结合几家酒店的评论

我已经导入它并通过tm包提供的转换实现了一些数据清理任务。

然后，当我使用以下脚本创建文档术语矩阵时：

DocumentTermMatrix(tm_map(reviewc, PlainTextDocument))

我得到的是一个没有任何单词但没有任何意义的字符的矩阵：

inspect(try[1:5, 200:500])
<<DocumentTermMatrix (documents: 5, terms: 301)>>
Non-/sparse entries: 0/1505
Sparsity           : 100%
Maximal term length: 25
Weighting          : term frequency (tf)

              Terms
Docs              â€œextensiveâ€\u009d    â€œextraâ€\u009d â€œfinest    â€œfreeâ€\u009d    â€œfromâ€\u009d    â€œfunkyâ€\u009d    â€œgoodâ€\u009d â€œhalf
  character(0)               0           0         0          0          0           0          0       0
  character(0)               0           0         0          0          0           0          0       0
  character(0)               0           0         0          0          0           0          0       0
  character(0)               0           0         0          0          0           0          0       0
  character(0)               0           0         0          0          0           0          0       0

任何人都知道我该怎么做才能避免这个错误？

提前感谢您的帮助

干杯！

Answer 1

library(tm) 
library(SnowballC) 
library(ggplot2)
library(FactoMineR)
library(RColorBrewer)
library(ape)
library(wordcloud)
library(stringr)

beijing_review <- read.csv("~/Downloads/beijing_review.csv", sep=";", comment.char="#")

# Remove this words
cleanwords = c("germany","alemania","bravcger", "\U0001f604\U0001f60a\U0001f44d\U0001f44d") ## Remove words

tryTolower = function(x)
{
  y = NA
  try_error = tryCatch(tolower(x), error=function(e) e)
  if (!inherits(try_error, "error"))
    y = tolower(x)
  return(y)
}

clean.text = function(x)
{
  # tolower
  x = tryTolower(x)
  # remove rt
  x = gsub("rt ", "", x)
  # remove at
  x = gsub("@\\w+", "", x)
  # remove punctuation
  x = gsub("[[:punct:]]", "", x)
  # remove numbers
  x = gsub("[[:digit:]]", "", x)
  # remove links http
  x = gsub("http\\w+", "", x)
  # remove tabs
  x = gsub("[ |\t]{2,}", "", x)
  # remove blank spaces at the beginning
  x = gsub("^ ", "", x)
  # remove blank spaces at the end
  x = gsub(" $", "", x)
  x = str_replace_all(x, "[^[:alnum:]]", " ")
  #return(x)
}


texto_c = clean.text(beijing_review$text) # Get column text 
texto_ac= paste(texto_c, collapse=" ")

rmNonAlphabet <- function(str) {
  words <- unlist(strsplit(str, " "))
  in.alphabet <- grep(words, pattern = "[a-z]", ignore.case = T)
  nice.str <- paste(words[in.alphabet], collapse = " ")
  nice.str
}

texto_ac = rmNonAlphabet(texto_ac)

busca_corpus = Corpus(VectorSource(texto_ac))

tdm = TermDocumentMatrix(busca_corpus,
                         control = list(removePunctuation = TRUE,
                                        stopwords = c(cleanwords,stopwords("english"),stopwords("spanish"),stopwords("portuguese"),cleanwords),
                                        removeNumbers = TRUE, tryTolower = TRUE, stopwords=TRUE))

m = as.matrix(tdm) 

palavras_freqs = sort(rowSums(m), decreasing=TRUE) # Contagem das palavras e ordenação 

dm= data.frame(word=names(palavras_freqs), freq=palavras_freqs)

dtm = DocumentTermMatrix(busca_corpus)

dtm_matrix = as.matrix(dtm)

top_palavras = head(palavras_freqs, 30) # nesse caso 10 usuários que mais tweetaram

barplot(top_palavras, border=NA, las=1, main="30 Top Words",  xlab="# of Rep", cex.main=1, horiz=TRUE, cex.names=0.65, axis.lty=1) 

# Plot WordCloud - Max word =100 and Freq >= 50
wordcloud(dm$word, dm$freq, random.order=FALSE,  min.freq=50,colors=brewer.pal(8, "Dark2"), max.words = 100)

文本挖掘 - 创建文档文本矩阵时出错（“扩展”）

1 个答案: