Spam Filter

Introduction

In this assignment, spam emails and non-spam (ham) emails are used to create a model that will determine if other emails are spam or not. To do this, hundreds of emails already characterized as spam or ham were labelled. A portion of the data frame was used as a training set, and the other portion was used as a testing set. The training data was used to make the model. The testing data was used to test the accuracy of the model.

Load libraries

library(tidyr)
library(tibble)
library(dplyr)

## 
## Attaching package: 'dplyr'

## The following objects are masked from 'package:stats':
## 
##     filter, lag

## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union

library(quanteda)

## Warning: package 'quanteda' was built under R version 4.3.2

## Warning in .recacheSubclasses(def@className, def, env): undefined subclass
## "ndiMatrix" of class "replValueSp"; definition not updated

## Warning in .recacheSubclasses(def@className, def, env): undefined subclass
## "pcorMatrix" of class "replValueSp"; definition not updated

## Package version: 3.3.1
## Unicode version: 13.0
## ICU version: 69.1

## Parallel computing: 12 of 12 threads used.

## See https://quanteda.io for tutorials and examples.

library(quanteda.textmodels)

## Warning: package 'quanteda.textmodels' was built under R version 4.3.2

library(caret)

## Loading required package: ggplot2

## Loading required package: lattice

library(readr)
library(caTools)

## Warning: package 'caTools' was built under R version 4.3.2

library(randomForest)

## randomForest 4.7-1.1

## Type rfNews() to see new features/changes/bug fixes.

## 
## Attaching package: 'randomForest'

## The following object is masked from 'package:ggplot2':
## 
##     margin

## The following object is masked from 'package:dplyr':
## 
##     combine

library(imputeTS)

## Warning: package 'imputeTS' was built under R version 4.3.2

## Registered S3 method overwritten by 'quantmod':
##   method            from
##   as.zoo.data.frame zoo

library(rpart)
library(purrr)

## 
## Attaching package: 'purrr'

## The following object is masked from 'package:caret':
## 
##     lift

library(rpart.plot)

## Warning: package 'rpart.plot' was built under R version 4.3.2

library(ROCR)

## Warning: package 'ROCR' was built under R version 4.3.2

First Attempt: Import the spam and non-spam files

When replicating this data, download the files found here:

https://spamassassin.apache.org/old/publiccorpus/20021010_easy_ham.tar.bz2 https://spamassassin.apache.org/old/publiccorpus/20021010_spam.tar.bz2

The files were too large to upload to Git Hub and use directly from this Rmd file.

Each step is explained in the comments of the R chunk. In the data frames, each column has each line of the emails.

# List all the files needed
not_spam_files <- list.files("spamham/easy_ham")
#not_spam_files <- not_spam_files[95]
spam_files <- list.files("spamham/spam")


# This is a for loop to get the words of all the ham files
# First, it gets the words from each file
# Each column is one file with all its words
# Each row shows a different line of each email
num <- 1
nonSpamEmails <- as.data.frame(col1 <- c(1:3100))
for (file in not_spam_files)
{
  del1 <- read.delim(paste0("spamham/easy_ham/", file), row.names = NULL)
  names(del1)[1] <- "one"
  
  if (ncol(del1) > 1)
  {
    names(del1)[2] <- "two"
    filename <- paste0("file", num)
    del1 <- unite(del1, filename, c(one, two))
  }
  else
  {
    names(del1)[1] <- "filename"
  }
  
  
  diffNum <- 3100 - nrow(del1)
  del2 <- data.frame(filename = c(rep(NA, diffNum)))
  del3 <- rbind(del1, del2)
  nonSpamEmails <- cbind(nonSpamEmails, del3)
  colnames(nonSpamEmails)[ncol(nonSpamEmails)] <- paste0("file", num)
  
  num <- num + 1
}

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

nonSpamEmails <- nonSpamEmails[,-1]

# This is a for loop to get the words of all the spam files
# First, it gets the words from each file
# Each column is one file with all its words
# Each row shows a different line of each email
num <- 1
spamEmails <- as.data.frame(col1 <- c(1:3100))
for (file in spam_files)
{
  del1 <- read.delim(paste0("spamham/spam/", file), row.names = NULL)
  names(del1)[1] <- "one"
  
  if (ncol(del1) > 1)
  {
    names(del1)[2] <- "two"
    filename <- paste0("file", num)
    del1 <- unite(del1, filename, c(one, two))
  }
  else
  {
    names(del1)[1] <- "filename"
  }
  
  
  diffNum <- 3100 - nrow(del1)
  del2 <- data.frame(filename = c(rep(NA, diffNum)))
  del3 <- rbind(del1, del2)
  spamEmails <- cbind(spamEmails, del3)
  colnames(spamEmails)[ncol(spamEmails)] <- paste0("file", num)
  
  num <- num + 1
}

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

## Warning in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
## : EOF within quoted string

spamEmails <- spamEmails[,-1]

Split into train and test data

The training data uses 75% of the spam and 75% of the ham emails. The testing data uses 25% of the spam and 25% of the ham emails. The column named spam is changed to 1 for spam and 0 for ham.

train <- c(spamEmails[,1:375], nonSpamEmails[,1:1913]) # This is the first 75% of the data
test <- c(spamEmails[,376:500], nonSpamEmails[,1913:2551]) # This is the last 25% of the data

train <- dfm(as.character(train))

## Warning: 'dfm.character()' is deprecated. Use 'tokens()' first.

test <- dfm(as.character(test))

## Warning: 'dfm.character()' is deprecated. Use 'tokens()' first.

trainClass <- c(rep(1, 375), rep(0, 1913))
testClass <- c(rep(1, 125), rep(0, 639))

Testing a model

This model did not work, so this model is fully commented. I included this model and others to show what failed.

The model did not work because the predictions did not make sense. They were between 1 and 2 instead of 0 and 1, so it is unclear what the predictions were showing.

model3 <- textmodel_wordscores(train, as.numeric(factor(trainClass)))
# Yes = 2
# No = 1
predictions <- predict(model3, newdata = test)

## Warning: 14947 features in newdata not used in prediction.

summary(predictions)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##   1.155   1.156   1.156   1.163   1.157   1.332

Second Attempt: Import the spam and non-spam files

This section formats the data differently because the data needed to be in a different format for the models I wanted to try. In this new format, each column is a word that appears in at least one of the emails, and each row is a file. The cells that line up with a word and file show the count of that word in that file.

# Get lists of all files needed
not_spam_files <- list.files("spamham/easy_ham")
spam_files <- list.files("spamham/spam")

notSpamFrequencies <- as.data.frame(c())
spamFrequencies <- as.data.frame(c())

# For loop to count each word in each email (non-spam)
for (file in not_spam_files)
{
  lines <- readLines(paste0("spamham/easy_ham/", file))
  lines2 <- paste(lines , collapse =" ")
  lines2 <- iconv(lines2, from = "ISO-8859-1", to = "UTF-8")
  words <- as.data.frame(strsplit(lines2, " "))
  
  words[words == ''] <- NA
  words[words == '...'] <- NA
  words <-
    words |> 
    na.omit()
  
  colnames(words)[1] = "one"
    
    # Count each word
  num <- 1
  for (myWord in words$one)
  {
    total <- 
      words |> 
      filter(one == myWord) |> 
      count() |> 
      as.integer()
    words$count[num] <- total
    num <- num + 1
  }
  
  words <- distinct(words)
  
  testing <- words |> pivot_wider(names_from = "one", values_from = "count")
  
  notSpamFrequencies <- bind_rows(notSpamFrequencies, testing) 
}

# For loop to count each word in each email (spam)
for (file in spam_files)
{
  lines <- readLines(paste0("spamham/spam/", file))
  lines2 <- paste(lines , collapse =" ")
  lines2 <- iconv(lines2, from = "ISO-8859-1", to = "UTF-8")
  words <- as.data.frame(strsplit(lines2, " "))
  
  words[words == ''] <- NA
  words[words == '...'] <- NA
  words <-
    words |> 
    na.omit()
  
  colnames(words)[1] = "one"
    
    # Count each word
  num <- 1
  for (myWord in words$one)
  {
    total <- 
      words |> 
      filter(one == myWord) |> 
      count() |> 
      as.integer()
    words$count[num] <- total
    num <- num + 1
  }
  
  words <- distinct(words)
  
  testing <- words |> pivot_wider(names_from = "one", values_from = "count")
  
  spamFrequencies <- bind_rows(spamFrequencies, testing)
}

## New names:
## New names:
## New names:
## New names:
## New names:
## New names:
## • `...12304` -> `...75`

## Warning in readLines(paste0("spamham/spam/", file)): incomplete final line
## found on 'spamham/spam/0143.260a940290dcb61f9327b224a368d4af'

## New names:
## New names:
## New names:
## New names:
## New names:
## New names:
## New names:
## New names:
## New names:
## New names:
## New names:
## New names:
## New names:
## New names:
## New names:
## New names:
## • `else......11` -> `else...`

Testing New Models

In this section, I showed the model that worked the best. None of the others worked. The accuracy was still very low at around 58%. However, the other models either did not work because of space limits or they resulted in horrible accuracy (less than 1% accuracy). Those are in the last section.

Disclaimer: This sections takes a long time to run, but it does finish eventually. Be patient, and it will work at some point.

# In this section, we want to create a model for new emails

# Add spam to the end
notSpamFrequencies$spam <- c(rep(0, nrow(notSpamFrequencies)))
spamFrequencies$spam <- c(rep(1, nrow(spamFrequencies)))

allFrequencies <- bind_rows(notSpamFrequencies, spamFrequencies)

allFrequencies$spam <- as.factor(allFrequencies$spam)

set.seed(1234)

spl = sample.split(allFrequencies$spam, 0.7)

train <- subset(allFrequencies, spl == TRUE)
test <- subset(allFrequencies, spl == FALSE)



eqnames <- names(train)[names(train) %in% names(test)]
train2 <- train[eqnames]
test2 <- test[eqnames]
  
train <- na_replace(train, 0)
train2 <- subset(train, select = -c(spam))
rf_classifier <- randomForest(x = train2[1500:2000,],
                          y = train$spam[1500:2000],
                          ntree = 100)

rf_classifier

## 
## Call:
##  randomForest(x = train2[1500:2000, ], y = train$spam[1500:2000],      ntree = 100) 
##                Type of random forest: classification
##                      Number of trees: 100
## No. of variables tried at each split: 374
## 
##         OOB estimate of  error rate: 0%
## Confusion matrix:
##     0   1 class.error
## 0 287   0           0
## 1   0 214           0

test <- na_replace(test, 0)


test2 <- subset(test, select = -c(spam))

rf_pred <- predict(rf_classifier, newdata = test2[380:880,])

confusionMatrix(table(rf_pred,test$spam[380:880]))

## Confusion Matrix and Statistics
## 
##        
## rf_pred   0   1
##       0 197   0
##       1 189 115
##                                           
##                Accuracy : 0.6228          
##                  95% CI : (0.5787, 0.6654)
##     No Information Rate : 0.7705          
##     P-Value [Acc > NIR] : 1               
##                                           
##                   Kappa : 0.3236          
##                                           
##  Mcnemar's Test P-Value : <2e-16          
##                                           
##             Sensitivity : 0.5104          
##             Specificity : 1.0000          
##          Pos Pred Value : 1.0000          
##          Neg Pred Value : 0.3783          
##              Prevalence : 0.7705          
##          Detection Rate : 0.3932          
##    Detection Prevalence : 0.3932          
##       Balanced Accuracy : 0.7552          
##                                           
##        'Positive' Class : 0               
##

Section 3:

In this section, I show the code for other methods I tried. These methods did not work for a few reasons. Mainly, the data frames were too large with too many rows and columns for proper analysis. R Studio could not handle the amount of data involved. I commented out all the sections, but you can uncomment them to test if you would like to see.

#spamCART <- rpart(spam~., data=train, method="class")


#train |> 
#    keep(~ isTRUE(any((. >= 50)((spam == 1) | (spam == 0)))))
#train |> keep(~ (isTRUE(any(. >= 50) & train$spam < 2)))

#train |> keep(c(~ isTRUE(any(c((. <= 2), (. >= 50))))), spam)

#smallTrain <- train |>
#  keep(~ isTRUE(any(. >= 50)))
#spamTrain <- train$spam

#over50 <- merge(spamTrain, smallTrain, 'row.names')
#over50 <- na.omit(over50)


#over50spamCART <- rpart(x~., data=over50, method="class")
#prp(over50spamCART)
#predTrainCART = predict(over50spamCART)[,2]
#predictionTrainCART = prediction(as.numeric(predTrainCART), over50$spam)

#as.numeric(performance(predictionTrainCART, "auc")@y.values)

#over50spamRF <- randomForest(x~., data=over50, na.action=na.omit)


#over50 <- over50[,-1]
#over50$x <- as.numeric(over50$x)
#spamLog = glm(x~., data=over50, family="binomial", na.action = na.omit)
#predTrainLog = predict(spamLog, type="response")

Sources:

I used all these sources. Some of them I just glanced at. Some of them I used to create the models. The first two websites helped a lot.