R语言作业四

R语言读取文件报错之一：incomplete final line found by readTableHeader on ‘xxxx.txt‘

停用词表下载链接

遇到的问题及解决：
载入了名字空间‘ellipsis’ 0.3.1，但需要的是＞= 0.3.2

频数提取代码：

install.packages('dplyr')
install.packages('jiebaR')
library(dplyr)
setwd('E:\\a乱七八糟\\r语言作业\\资料')  #将工作文件夹转换到当前目录
stopwords <- readtext::readtext("stopwords.txt") %>% 
    as.character() %>% 
    stringr::str_split('\n') %>% 
    unlist()
    #tolower(stopwords)
 
#显示前50个停用词
head(stopwords, n=50)
library(jiebaR)
#告诉worker停用词表的位置
tokenizer <- worker(stop_word = 'stopwords.txt')
 
#读取三体.txt为字符串
text <- readtext::readtext("debate.txt") %>% as.character()
 
#分词
words <- segment(text, tokenizer)
 
#显示分词结果的前20个词
head(words, n=20)
new_words <- c()
for (word in words) {
    if (!word %in% stopwords){
        new_words <- c(new_words, word)
    }
}
head(new_words)
wordfreqs <- jiebaR::freq(new_words)
wordfreqs <- dplyr::arrange(wordfreqs, -freq)
#wordfreqs
library('wordcloud')
wordcloud(wordfreqs$char , wordfreqs$freq, colors = rainbow(300), random.order=F)

关键词代码

install.packages('pacman')
library(pacman)
library(dplyr)
p_load(tidyverse,tidytext,textrank,rio,jiebaR)
p_load(cidian)
wordfreqs %>%  mutate(id = 1:1) -> f_table  
f_table %>%  bind_tf_idf(term = char,document = id,n = freq) -> tf_idf_table

tf_idf_table %>% 
  group_by(id) %>% 
top_n(3,tf) %>% 
 ungroup() -> top3
wordcloud(top3$char , top3$freq, colors = rainbow(30), random.order=F)

参考资料：
R语言自然语言处理：关键词提取（TF-IDF）
情感分析代码

install.packages("tidyr")
library(tidyr)
get_sentiments("bing")
 wordfreqs %>%  mutate(id = 1:n()) -> wordfreqs
woman_D<-inner_join(wordfreqs,get_sentiments("bing"), by = c("char" = "word"))
woman_E<-count(woman_D,index=id%/%40,sentiment)
 woman_F<-spread(woman_E,sentiment,n)
woman_G<-mutate(woman_F,sentiment=positive-negative)
ggplot(woman_G,aes(index,sentiment))+geom_col()