术语对和短语的tm_term_score错误

Question

尝试使用tm_term_score为行中的文本对术语短语或术语对评分为1或0会得到以下错误：

c1 <- Corpus(VectorSource(r1))
inspect(c1)
inspect(c1[2])
cleanR1 <- tm_map(c1, removeWords, stopwords('english'))
cleanR1 <- tm_map(cleanR1, content_transformer(tolower))
cleanR1 <- tm_map(cleanR1, removeWords, c("re", "ve", "ahha"))   
cleanR1 <- tm_map(cleanR1, removePunctuation)
cleanR1 <- tm_map(cleanR1, removeSpecialChars)
cleanR1 <- tm_map(cleanR1, removeNumbers)
cleanR1 <- tm_map(cleanR1,removeWords,stopwords('en'))
cleanR1 <- tm_map(cleanR1,stripWhitespace)
cleanR1 <- tm_map(cleanR1,stemDocument)
cleanR1 <- tm_map(cleanR1, PlainTextDocument)

tm_term_score(cleanR1, terms, FUN = slam::row_sum) 

UseMethod（“ tm_term_score”，x）中的错误：没有适用于'tm_term_score'的适用方法应用于类“ c（'VCorpus'，'Corpus'）”的对象

str（terms）chr [1:61]“帮助学习”“帮助成长”。

Answer 1

你需要传递一个PlainTextDocument ，通过返回的词频termFreq ，或TermDocumentMatrix到的第一个参数tm_term_score而不是语料库。

使用术语文档矩阵的示例：

tdm <- TermDocumentMatrix(cleanR1)
tm_term_score(tdm, terms, FUN = slam::row_sum)

请参阅文档?tm::tm_term_score