尝试创建一个 R Shiny 应用程序来抓取 Twitter 并创建一个词云

Question

我以前从来没有用过 Shiny，所以如果这是一个非常愚蠢的问题，我深表歉意。我正在尝试制作一个闪亮的应用程序，您可以在其中输入将从 Twitter 中提取的搜索词并创建词云。我觉得我快到了，但它不起作用，老实说，我真的不知道我在做什么。我正在尝试自学 Shiny，但我也知道像这样的论坛对学习非常有用。

library(tm)
library(wordcloud)
library(memoise)



ui <- fluidPage(
# Application title
titlePanel("Word Cloud"),

sidebarLayout(
# Sidebar with a slider and selection inputs
sidebarPanel(
textInput("selection", "Input your search term:",
              ""),
actionButton("update", "Change"),
hr(),
sliderInput("freq",
              "Minimum Frequency:",
              min = 1,  max = 50, value = 15),
  sliderInput("max",
              "Maximum Number of Words:",
              min = 1,  max = 300,  value = 100)
),

# Show Word Cloud
mainPanel(
  plotOutput("plot")
)
)
)
#Define server logic

server <- function(input, output, session) {
 # Define a reactive expression for the document term matrix
  terms <- reactive({
consumer_key <- "XXXX"
consumer_secret <- "XXXX"
access_token <- "XXXX"
access_secret <- "XXXX"
#Here we are creating the "handshake" with Twitter
setup_twitter_oauth(consumer_key= consumer_key, consumer_secret= 
consumer_secret,access_token= access_token, access_secret= access_secret)

#Once you have created your handshake, you can start searching for tweets
#Note that if you select a common term like "Atlanta" you will generate a lot 
of Tweets quickly
#But if you select an esoteric term like "heteroscedasticity", it might take 
a while to get any
tw<-searchTwitter("selection", n=1000, lang='en', resultType = "recent")
# Using "memoise" to automatically cache the results
getTermMatrix <- memoise(function(tw) {
  text <- readLines(sprintf(tw),
                    encoding="UTF-8")

  myCorpus = Corpus(VectorSource(text))
  myCorpus = tm_map(myCorpus, content_transformer(tolower))
  myCorpus = tm_map(myCorpus, removePunctuation)
  myCorpus = tm_map(myCorpus, removeNumbers)
  myCorpus = tm_map(myCorpus, removeWords,
                    c(stopwords("SMART"), "thy", "thou", "thee", "the", 
"and", "but"))

  myDTM = TermDocumentMatrix(myCorpus,
                             control = list(minWordLength = 1))

  m = as.matrix(myDTM)

  sort(rowSums(m), decreasing = TRUE)
})

# Change when the "update" button is pressed...
input$update
# ...but not for anything else
isolate({
  withProgress({
    setProgress(message = "Processing corpus...")
    getTermMatrix(input$selection)

  })


})

 })
# Make the wordcloud drawing predictable during a session
 wordcloud_rep <- reactive({

 v <- terms()
 wordcloud_rep(names(v), v, scale=c(4,0.5),
            min.freq = input$freq, max.words=input$max,
            colors=brewer.pal(8, "Dark2"))

})
}

# Run the application 
shinyApp(ui = ui, server = server)

感谢任何试图提供帮助的人！！

编辑：啊，不好意思，没说清楚哪里出了问题！到目前为止，它打开了一个包含我想要的所有输入框的应用程序，但输入搜索词似乎实际上没有做任何事情。它只是永远加载。没有错误。

Answer 1

那么你的问题是什么？还是您寻求一般帮助？我看到的第一件事是你正在存储秘密信息 (consumer_key <- "XXXX",consumer_secret <- "XXXX", access_token <- "XXXX"...) 在您的服务器逻辑中。最好将它们存储在您的 .Renviron 中并使用 Sys.getenv("token").

调用它们

Answer 2

嘿抱歉第一次回答...,

该应用程序使用以下代码，我认为您应该考虑一个基本的闪亮教程来完全理解所有内容，即输出以与输入相同的方式连接到 ui --> server: output$plot <- renderPlot({...}) 和 ui: plotOutput("plot")。我使用 observeEvent 检查更新按钮是否被点击。最大值的滑块输入。 Words 也在工作，但不是最小的。频率，因为我不知道你在用它做什么。我也只是做了一个基本的单词清理。希望对您有所帮助：

library(tm)
library(wordcloud)
library(memoise)
# new libs
library(twitteR)
library(tidyverse)
library(tidytext)


ui <- fluidPage(
    # Application title
    titlePanel("Word Cloud"),

    sidebarLayout(
        # Sidebar with a slider and selection inputs
        sidebarPanel(
            textInput("selection", "Input your search term:",
                      ""),
            actionButton("update", "Change"),
            hr(),
            sliderInput("freq",
                        "Minimum Frequency:",
                        min = 1,  max = 50, value = 15),
            sliderInput("max",
                        "Maximum Number of Words:",
                        min = 1,  max = 300,  value = 100)
        ),

        # Show Word Cloud
        mainPanel(
            plotOutput("plot")
        )
    )
)
#Define server logic

api_key <- Sys.getenv("twitter_api")
api_secret  <- Sys.getenv("twitter_secret")
access_token <- Sys.getenv("twitter_access_token")
access_secret <- Sys.getenv("twitter_access_secret")

server <- function(input, output, session) {
    tweets_clean <- reactiveValues(df = NULL)
    # Define a reactive expression for the document term matri
        #Here we are creating the "handshake" with Twitter
        setup_twitter_oauth(access_token = access_token ,access_secret = access_secret,
                            consumer_key = api_key,consumer_secret = api_secret )

    observeEvent(input$update,{

        tw <- searchTwitter(input$selection, n=input$max, lang='en', resultType = "recent")
        # tweets to df so we could use tidytext
        df <- twListToDF(tw)
        # use dplyr and tidytext to clean your tweets
        tweets_clean$df <- df %>%
            dplyr::select(text) %>%
            tidytext::unnest_tokens(word, text) %>%
            count(word, sort = TRUE) 
    })
    output$plot <- renderPlot({
        # plot it
        if(is.null(tweets_clean$df)){
            NULL
        } else{
            wordcloud(tweets_clean$df$word, tweets_clean$df$n)
        }
    })

}

# Run the application 
shinyApp(ui = ui, server = server)`enter code here`

尝试创建一个 R Shiny 应用程序来抓取 Twitter 并创建一个词云

Trying to create an R Shiny app that will crawl Twitter and create a word cloud

twitter

r

word

scrape

shiny