R/comparison.R

In zahrakhoshmanesh/FinalProjectSTAT585: Sentiment Analysis of Reviews

#' visualize performance of the algorithm
#'
#' @param x input is a dataframe for bag of word file in which columns are the terms and row are binary variable 1 if that term exist in that data instance
#' @return visualization of trained algrithms (for classifiying textual data), comparision and select the best!
#' @author Atousa Zarindast
#' @export
#' @examples
#' \dontrun{
#' library(SentiAnalyzer)
#' csv_data <- read.csv(system.file(package = "SentiAnalyzer", "extdata/testing.csv"))
#' comparison <- comparison(csv_data)
#' }
#' 
comparison <- function(x) {
  conf <- NULL
  
  if (ncol(x) > 1) {
    df <- BuildPrediction(x)
  } else {
    assertthat::assert_that(x%>%purrr::map(.f=length)==5)
    df <- x
  }
  
  accuracy_f <- function(x) {
    (x$table[1] + x$table[2, 2]) / (x$table[1] + x$table[1, 2] + x$table[2, 1] + x$table[2, 2])
  }
  precision_f <- function(x) {
    (x$table[1]) / (x$table[1] + x$table[1, 2])
  }
  recall_f <- function(x) {
    (x$table[1]) / (x$table[1] + x$table[2, 1])
  }
  f1score <- function(x) {
    2 * ((x$table[1]) / (x$table[1] + x$table[1, 2])) * ((x$table[1]) / (x$table[1] + x$table[2, 1])) / (((x$table[1]) / (x$table[1] + x$table[1, 2])) + (x$table[1]) / (x$table[1] + x$table[2, 1]))
  }
  
  
  
  df <- df %>% dplyr::mutate(
    accuracy = purrr::map(.x = conf, .f = accuracy_f),
    precision = purrr::map(.x = conf, .f = precision_f),
    recall = purrr::map(.x = conf, .f = recall_f),
    f1score = purrr::map(.x = conf, .f = f1score)
  )
  
  
  funm <- function(x) {
    as.data.frame(unlist(x))
  }
  df <- df[-1] %>% 
    purrr::map(.f = funm) %>% 
    as.data.frame() %>% 
    dplyr::mutate(method = c("gbm", "KNN", "NB", "RandomForest", "SVM"))
  names(df) <- (c("Accuracy", "Precision", "Recall", "F1", "Method"))
  
  return(df)
}