我使用它来获取一个词的频率：
library(tm)
df <- c('Dummy text: Its function as a filler or as a tool for comparing the visual impression of different typefaces Dummy text is text that is used in the publishing industry or by web designers to occupy the space which will later be filled with 'real' content. This is required when, for example, the final text is not yet available. Dummy text is also known as 'fill text'. It is said that song composers of the past used dummy texts as lyrics when writing melodies in order to have a 'ready-made' text to sing with the melody. Dummy texts have been in use by typesetters since the 16th century. work in team The usefulness of nonsensical content Dummy text is also used to demonstrate the appearance of different typefaces and layouts, and in general the content of dummy text is nonsensical. Due to its widespread use as filler text for layouts, non-readability is of great importance: human perception is tuned to recognize certain patterns and repetitions in texts. If the distribution of letters and 'words' is random, the reader will not be distracted from making a neutral judgement on the visual impact and readability of the typefaces (typography), or the distribution of text on the page (layout or type area). For this reason, dummy text usually consists of a more or less random series of words or syllables. This prevents repetitive patterns from impairing the overall visual impression and facilitates the comparison of different typefaces. Furthermore, it is advantageous when the dummy text is relatively realistic so that the layout impression of the final publication is not compromised. work in team Incomprehensibility or readability? That is the question. The most well-known dummy text is the 'Lorem Ipsum', which is said to have originated in the 16th century. Lorem Ipsum is composed in a pseudo-Latin language which more or less corresponds to 'proper' Latin. It contains a series of real Latin words. This ancient dummy text is also incomprehensible, but it imitates the rhythm of most European languages in Latin script. The advantage of its Latin origin and the relative meaninglessness of Lorum Ipsum is that the text does not attract attention to itself or distract the viewer's attention from the layout. One disadvantage of Lorum Ipsum is that in Latin certain letters appear more frequently than others - which creates a distinct visual impression. Moreover, in Latin only words at the beginning of sentences are capitalized; this means that Lorem Ipsum cannot accurately represent, for example, German, in which all nouns are capitalized. Thus, Lorem Ipsum has only limited suitability as a visual filler for German texts. If the fill text is intended to illustrate the characteristics of different typefaces, it sometimes makes sense to select texts containing the various letters and symbols specific to the output language. work in team There is now an abundance of readable dummy texts. These are usually used when a text is required purely to fill a space. These alternatives to the classic Lorem Ipsum texts are often amusing and tell short, funny or nonsensical stories. Origin and meaning of the Lorem Ipsum text According to most sources, Lorum Ipsum can be traced back to a text composed by Cicero in 45 BC. Allegedly, a Latin scholar established the origin of the text by compiling all the instances of the unusual word 'consectetur' he could find and eventually recognized it as a passage from 'De finibus bonorum et malorum' (On the extremes of Good and Evil) by Cicero, a book that was very popular in the Middle Ages: "Neque porro quisquam est, qui dolorem ipsum, quia dolor sit, amet, consectetur, adipisci velit... " (Approximately: "Nor is there anyone who loves or pursues or desires to obtain pain of itself, because it is pain..."). A typical Lorem Ipsum text goes like this: "Lorem ipsum dolor sit amet, consectetur adipisici elit, sed eiusmod tempor incidunt ut labore et dolore magna aliqua. Ut enim ad minim ..." working in team It seems that only fragments of the original text remain in the Lorem Ipsum texts used today. One may speculate that over the course of time certain letters were added or deleted at various positions within the text. This might also explain why one can now find slightly different versions. Due to the age of the Lorem Ipsum text there are no copyright issues to contend with. work in team In the 1960s, the text suddenly became known beyond the professional circle of typesetters and layout designers when it was used for Letraset sheets (adhesive letters on transparent film, popular until the 1980s) Versions of the text were subsequently included in DTP programmes such as PageMaker etc. work in team Automatic recognition of Lorem Ipsum during the pre-press workflow The spread of computers and layout programmes thus made dummy text better known. While in earlier times several lines of the Lorem Ipsum text were repeated in the creation of dummy texts, today the full text of Cicero's 'De finibus' serves as the basis for many dummy text or Lorem Ipsum generators. Based on 'De finibus', these generators automatically create longer sections of the Lorem Ipsum text or various other filler texts. work in team The phrasal sequence of the Lorem Ipsum text is now so widespread and commonplace that many DTP programmes can generate dummy text using the starting sequence "Lorem ipsum". Fortunately, the phrase 'Lorem Ipsum' is now recognized by electronic pre-press systems and, when found, an alarm can be raised. This avoids a publication going to print with overlooked dummy text. Certain internet providers exploit the fact that fill text cannot be recognized by automatic search engines - meaningful information cannot be distinguished from meaningless: Target-generated dummy text mixed with a certain combination of search terms can lead to an increased frequency of visits by search machine users. As a consequence, advertising revenues, which rely on website 'hits', are increased.')
myStopwords <- c("will", "world")
review_source <- VectorSource(df)
corpus <- Corpus(review_source)
corpus <- Corpus(review_source)
corpus <- tm_map(corpus, content_transformer(tolower))
corpus <- tm_map(corpus, removePunctuation)
corpus <- tm_map(corpus, stripWhitespace)
corpus <- tm_map(corpus, removeWords, stopwords("english"))
corpus <- tm_map(corpus, removeWords, myStopwords)
corpus <- tm_map(corpus, stemDocument)
dtm <- DocumentTermMatrix(corpus)
dtm2 <- as.matrix(dtm)
frequency <- colSums(dtm2)
frequency <- sort(frequency, decreasing = TRUE)
head(frequency)
为了创建ngram，我使用here中的示例代码：
library(RWeka)
BigramTokenizer <- function(x) NGramTokenizer(x, Weka_control(min = 2, max = 2))
txtTdmBi <- TermDocumentMatrix(corpus, control = list(tokenize = BigramTokenizer))
我怎样才能看到txtTdmBi的结果，就像用单个单词一样？我还应该在哪里放置选项options(mc.cores=1)来将ngram改为bigram，trigram等？我在txtTdmBi中使用了它，但它给出了一个错误。
更新
我可以使用这个看到结果：
inspect(txtTdmBi)
但结果是单个单词。我想我应该有不同的文件。是否可以在同一份文件中制作？
答案 0 :(得分：1)
关于1克，2克等问题;这由参数X和Y控制，如：
Weka_control(min = X, max = Y)
举一些例子：
X=1 and Y=1--> Then you will get 1-gram
X=2 and Y=2--> Then you will get 2-gram
X=1 and Y=2--> Then you will get 1-gram and 2-grams