计算列总比例与所有列总数的比较

Question

我正在尝试计算列总数与所有列总数的比例。

我正在使用的文档术语矩阵非常大，导致我运行错误的任何测试基本上都会导致Rstudio崩溃。

到目前为止，这是我的工作代码：

randomSample = read.csv("randomSample2016.csv", stringsAsFactors = FALSE)
str(randomSample)
randomSample$tweet <- as.character(randomSample$tweet)

randomSample$tweetlength <- nchar(randomSample$tweet)

hist(randomSample$tweetlength)


library("tm")

## Use the tm library to construct a document-term matrix of term 
frequencies

randomSample_corpus <- Corpus(VectorSource(randomSample$tweet))
print(randomSample_corpus)

inspect(randomSample_corpus[1:3])

#clean up corpus
#make all letters lowercase
randomSample_corpus_clean <- tm_map(randomSample_corpus, tolower)

#Remove Numbers
randomSample_corpus_clean <- tm_map(randomSample_corpus_clean, 
removeNumbers)

#Remove punctuation
randomSample_corpus_clean <- tm_map(randomSample_corpus_clean, 
removePunctuation)

#Remove stop words
randomSample_corpus_clean <- tm_map(randomSample_corpus_clean, removeWords, 
stopwords())

#remove unneeded whitespace
randomSample_corpus_clean <- tm_map(randomSample_corpus_clean, 
stripWhitespace)

#Inspect cleaned corpus
 inspect(randomSample_corpus_clean[1:3])

#Create document term matrix 
randomSample_dtm <- DocumentTermMatrix(randomSample_corpus_clean)

#convert to R matrix
dtm2 <- as.matrix(randomSample_dtm)

#obtain individual word frequencies
frequency <- colSums(dtm2)

这允许我获取dtm2中所有单词的总频率，但是当我尝试为列总计（dtm2$newcolumn <- 0）添加新行时，我最终会使用疯狂的ram使用量。

Answer 1

那是因为你的################################################################################ # Get path of file selected in Finder ################################################################################ getsel() { osascript <<EOF tell application "Finder" to return POSIX path of first item of (get selection as alias list) EOF } 返回了一个命名数组。你想要的是

colSums