Shiny R App Correlation by Taking User Selected Variables

user_16564530 0
Viewed 295

I would like my shiny app to take the user's selected variables from my data set (called "CD" in this case) and I would like it to print two things: the correlation and a scatterplot.

to start, here's a head of the data:

 wins kills kdRatio killstreak losses headshots misses scorePerMinute
  <dbl> <dbl>   <dbl>      <dbl>  <dbl>     <dbl>  <dbl>          <dbl>
1     0     0    0             0      0         0      0             0 
2     0     0    0             0      0         0      0             0 
3     0    66    1.03          0      0        16      0             0 
4     3     2    0.4           0      0         0      0             0 
5     0     2    0.2           0      0         1      0             0 
6   684 27011    1.07         18     10      5113 305319           256.

the variables are named "variables:

variables = names(CD)

this is the progress I have on the ui side of things (for this tab at least):

tabPanel("Correlation Metrics",
                         headerPanel("Correlations Between Different COD Stats"),
                         mainPanel(
                           "Now I would like to show some correlation metrics to give a better understanding of which Call of Duty
                           stats really correlate with others. Please select the variables that you are interested and observe
                           which CoD stats correlate!",
                          
                           selectInput('var1', "First Variable", variables),
                           selectInput('var2', "Second Variable", variables, selected = variables[[2]]),
                           plotOutput('correlation')
                        
                         )

and this is the progress I have on the server side for this tab:

output$correlation<- renderPlot({
    x<-as.numeric('var1')
    y<-as.numeric('var2')
    plot(x, y)
  })

I know I am way off, but please help me fix this!

shiny correlation r
1 Answers

We may need to subset the column from the input. Based on the input data showed, columns are all numeric, so there is no need to do as.numeric again after extraction

output$correlation<- renderPlot({
    req(input$var1)
    req(input$var2)
    x <- CD[[input$var1]]
    y <- CD[[input$var2]]
    plot(x, y)
  })

-full code

library(shiny)

CD <- structure(list(wins = c(0L, 0L, 0L, 3L, 0L, 684L), kills = c(0L, 
                                                                   0L, 66L, 2L, 2L, 27011L), kdRatio = c(0, 0, 1.03, 0.4, 0.2, 1.07
                                                                   ), killstreak = c(0L, 0L, 0L, 0L, 0L, 18L), losses = c(0L, 0L, 
                                                                                                                          0L, 0L, 0L, 10L), headshots = c(0L, 0L, 16L, 0L, 1L, 5113L), 
                     misses = c(0L, 0L, 0L, 0L, 0L, 305319L), scorePerMinute = c(0, 
                                                                                 0, 0, 0, 0, 256)), class = "data.frame", row.names = c("1", 
                                                                                                                                        "2", "3", "4", "5", "6"))
variables = names(CD)
ui <- fluidPage(
  tabPanel("Correlation Metrics",
           headerPanel("Correlations Between Different COD Stats"),
           mainPanel(
             "Now I would like to show some correlation metrics to give a better understanding of which Call of Duty
                           stats really correlate with others. Please select the variables that you are interested and observe
                           which CoD stats correlate!",
             
             selectInput('var1', "First Variable", variables),
             selectInput('var2', "Second Variable", variables, selected = variables[[2]]),
             plotOutput('correlation')
             
           )
           
))

server <- function(input, output) {
  output$correlation<- renderPlot({
    req(input$var1)
    req(input$var2)
    x <- CD[[input$var1]]
    y <- CD[[input$var2]]
    plot(x, y)
  })
  
}

shinyApp(ui = ui, server = server)

-output

enter image description here

user_37322710
Related