For Machine Learning and Web Analytics blogpost http://markedmondson.me/intro-to-machine-learning-with-web-analytics-random-forests-and-k-means

ml_plot_randomForests.R

## function to get plot data format
getCompareTable <- function (test_data, prediction) {
  require(dplyr)
  
  ## plot real vs model bought Sku
  actual_freq <- table(model_data$boughtSku)
  predicted_freq <- table(prediction)
  
  actual_freq <- actual_freq[order(actual_freq)]
  predicted_freq <- predicted_freq[order(predicted_freq)]
  
  actual_freq_s <- data.frame(sku = names(actual_freq), 
                              actual = as.vector(actual_freq), 
                              stringsAsFactors = F)
  
  predicted_freq_s <- data.frame(sku = names(predicted_freq),
                                 predict = as.vector(predicted_freq),
                                 stringsAsFactors = F)
  
  actual_freq_s$actual <- unname(actual_freq_s$actual)
  predicted_freq_s$predict <- unname(predicted_freq_s$predict)
  
  compare <- dplyr::left_join(actual_freq_s, predicted_freq_s, by = "sku")
  compare
}

## use function to get plot data
compare <- getCompareTable(test, prediction)

## plot the predicted vs actual in test set
library(ggplot2)
library(reshape2)

compare_long <- melt(compare)
g <- ggplot(data = compare_long, aes(x=sku, y = value, colour = variable, group = variable)) + theme_bw()
g <- g + geom_bar(stat = "identity", position = "dodge", aes(fill=variable))
g