Skip to content
Permalink
Branch: master
Find file Copy path
Find file Copy path
Fetching contributors…
Cannot retrieve contributors at this time
2439 lines (2438 sloc) 363 KB
{
"cells": [
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"Loading required package: lattice\n",
"Loading required package: ggplot2\n",
"── Attaching packages ─────────────────────────────────────── tidyverse 1.2.1 ──\n",
"✔ tibble 2.0.1 ✔ purrr 0.3.0 \n",
"✔ tidyr 0.8.2 ✔ dplyr 0.8.0.1\n",
"✔ readr 1.3.1 ✔ stringr 1.4.0 \n",
"✔ tibble 2.0.1 ✔ forcats 0.4.0 \n",
"── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──\n",
"✖ dplyr::filter() masks stats::filter()\n",
"✖ dplyr::lag() masks stats::lag()\n",
"✖ purrr::lift() masks caret::lift()\n",
"\n",
"Attaching package: ‘magrittr’\n",
"\n",
"The following object is masked from ‘package:purrr’:\n",
"\n",
" set_names\n",
"\n",
"The following object is masked from ‘package:tidyr’:\n",
"\n",
" extract\n",
"\n"
]
}
],
"source": [
"library(caret)\n",
"library(tidyverse)\n",
"library(magrittr)"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [],
"source": [
"#The German Credit Dataset - https://archive.ics.uci.edu/ml/datasets/default+of+credit+card+clients\n",
"#Input data\n",
"pathgerman <- (\"https://raw.githubusercontent.com/longtng/frauddetectionproject/master/german_credit.csv\")\n",
"german <- read.csv(pathgerman,header = T)"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {
"scrolled": true
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"'data.frame':\t1000 obs. of 62 variables:\n",
" $ Duration : int 6 48 12 42 24 36 24 36 12 30 ...\n",
" $ Amount : int 1169 5951 2096 7882 4870 9055 2835 6948 3059 5234 ...\n",
" $ InstallmentRatePercentage : int 4 2 2 2 3 2 3 2 2 4 ...\n",
" $ ResidenceDuration : int 4 2 3 4 4 4 4 2 4 2 ...\n",
" $ Age : int 67 22 49 45 53 35 53 35 61 28 ...\n",
" $ NumberExistingCredits : int 2 1 1 1 2 1 1 1 1 2 ...\n",
" $ NumberPeopleMaintenance : int 1 1 2 2 2 2 1 1 1 1 ...\n",
" $ Telephone : int 0 1 1 1 1 0 1 0 1 1 ...\n",
" $ ForeignWorker : int 1 1 1 1 1 1 1 1 1 1 ...\n",
" $ Class : Factor w/ 2 levels \"Bad\",\"Good\": 2 1 2 2 1 2 2 2 2 1 ...\n",
" $ CheckingAccountStatus.lt.0 : int 1 0 0 1 1 0 0 0 0 0 ...\n",
" $ CheckingAccountStatus.0.to.200 : int 0 1 0 0 0 0 0 1 0 1 ...\n",
" $ CheckingAccountStatus.gt.200 : int 0 0 0 0 0 0 0 0 0 0 ...\n",
" $ CheckingAccountStatus.none : int 0 0 1 0 0 1 1 0 1 0 ...\n",
" $ CreditHistory.NoCredit.AllPaid : int 0 0 0 0 0 0 0 0 0 0 ...\n",
" $ CreditHistory.ThisBank.AllPaid : int 0 0 0 0 0 0 0 0 0 0 ...\n",
" $ CreditHistory.PaidDuly : int 0 1 0 1 0 1 1 1 1 0 ...\n",
" $ CreditHistory.Delay : int 0 0 0 0 1 0 0 0 0 0 ...\n",
" $ CreditHistory.Critical : int 1 0 1 0 0 0 0 0 0 1 ...\n",
" $ Purpose.NewCar : int 0 0 0 0 1 0 0 0 0 1 ...\n",
" $ Purpose.UsedCar : int 0 0 0 0 0 0 0 1 0 0 ...\n",
" $ Purpose.Furniture.Equipment : int 0 0 0 1 0 0 1 0 0 0 ...\n",
" $ Purpose.Radio.Television : int 1 1 0 0 0 0 0 0 1 0 ...\n",
" $ Purpose.DomesticAppliance : int 0 0 0 0 0 0 0 0 0 0 ...\n",
" $ Purpose.Repairs : int 0 0 0 0 0 0 0 0 0 0 ...\n",
" $ Purpose.Education : int 0 0 1 0 0 1 0 0 0 0 ...\n",
" $ Purpose.Vacation : int 0 0 0 0 0 0 0 0 0 0 ...\n",
" $ Purpose.Retraining : int 0 0 0 0 0 0 0 0 0 0 ...\n",
" $ Purpose.Business : int 0 0 0 0 0 0 0 0 0 0 ...\n",
" $ Purpose.Other : int 0 0 0 0 0 0 0 0 0 0 ...\n",
" $ SavingsAccountBonds.lt.100 : int 0 1 1 1 1 0 0 1 0 1 ...\n",
" $ SavingsAccountBonds.100.to.500 : int 0 0 0 0 0 0 0 0 0 0 ...\n",
" $ SavingsAccountBonds.500.to.1000 : int 0 0 0 0 0 0 1 0 0 0 ...\n",
" $ SavingsAccountBonds.gt.1000 : int 0 0 0 0 0 0 0 0 1 0 ...\n",
" $ SavingsAccountBonds.Unknown : int 1 0 0 0 0 1 0 0 0 0 ...\n",
" $ EmploymentDuration.lt.1 : int 0 0 0 0 0 0 0 0 0 0 ...\n",
" $ EmploymentDuration.1.to.4 : int 0 1 0 0 1 1 0 1 0 0 ...\n",
" $ EmploymentDuration.4.to.7 : int 0 0 1 1 0 0 0 0 1 0 ...\n",
" $ EmploymentDuration.gt.7 : int 1 0 0 0 0 0 1 0 0 0 ...\n",
" $ EmploymentDuration.Unemployed : int 0 0 0 0 0 0 0 0 0 1 ...\n",
" $ Personal.Male.Divorced.Seperated : int 0 0 0 0 0 0 0 0 1 0 ...\n",
" $ Personal.Female.NotSingle : int 0 1 0 0 0 0 0 0 0 0 ...\n",
" $ Personal.Male.Single : int 1 0 1 1 1 1 1 1 0 0 ...\n",
" $ Personal.Male.Married.Widowed : int 0 0 0 0 0 0 0 0 0 1 ...\n",
" $ Personal.Female.Single : int 0 0 0 0 0 0 0 0 0 0 ...\n",
" $ OtherDebtorsGuarantors.None : int 1 1 1 0 1 1 1 1 1 1 ...\n",
" $ OtherDebtorsGuarantors.CoApplicant : int 0 0 0 0 0 0 0 0 0 0 ...\n",
" $ OtherDebtorsGuarantors.Guarantor : int 0 0 0 1 0 0 0 0 0 0 ...\n",
" $ Property.RealEstate : int 1 1 1 0 0 0 0 0 1 0 ...\n",
" $ Property.Insurance : int 0 0 0 1 0 0 1 0 0 0 ...\n",
" $ Property.CarOther : int 0 0 0 0 0 0 0 1 0 1 ...\n",
" $ Property.Unknown : int 0 0 0 0 1 1 0 0 0 0 ...\n",
" $ OtherInstallmentPlans.Bank : int 0 0 0 0 0 0 0 0 0 0 ...\n",
" $ OtherInstallmentPlans.Stores : int 0 0 0 0 0 0 0 0 0 0 ...\n",
" $ OtherInstallmentPlans.None : int 1 1 1 1 1 1 1 1 1 1 ...\n",
" $ Housing.Rent : int 0 0 0 0 0 0 0 1 0 0 ...\n",
" $ Housing.Own : int 1 1 1 0 0 0 1 0 1 1 ...\n",
" $ Housing.ForFree : int 0 0 0 1 1 1 0 0 0 0 ...\n",
" $ Job.UnemployedUnskilled : int 0 0 0 0 0 0 0 0 0 0 ...\n",
" $ Job.UnskilledResident : int 0 0 1 0 0 1 0 0 1 0 ...\n",
" $ Job.SkilledEmployee : int 1 1 0 1 1 0 1 0 0 0 ...\n",
" $ Job.Management.SelfEmp.HighlyQualified: int 0 0 0 0 0 0 0 1 0 1 ...\n"
]
}
],
"source": [
"german %>% str()"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [],
"source": [
"# Checking missing value for each column function\n",
"naratio <- function(x) {100*sum(is.na(x)) / length(x)}"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<table>\n",
"<thead><tr><th scope=col>Duration</th><th scope=col>Amount</th><th scope=col>InstallmentRatePercentage</th><th scope=col>ResidenceDuration</th><th scope=col>Age</th><th scope=col>NumberExistingCredits</th><th scope=col>NumberPeopleMaintenance</th><th scope=col>Telephone</th><th scope=col>ForeignWorker</th><th scope=col>Class</th><th scope=col>⋯</th><th scope=col>OtherInstallmentPlans.Bank</th><th scope=col>OtherInstallmentPlans.Stores</th><th scope=col>OtherInstallmentPlans.None</th><th scope=col>Housing.Rent</th><th scope=col>Housing.Own</th><th scope=col>Housing.ForFree</th><th scope=col>Job.UnemployedUnskilled</th><th scope=col>Job.UnskilledResident</th><th scope=col>Job.SkilledEmployee</th><th scope=col>Job.Management.SelfEmp.HighlyQualified</th></tr></thead>\n",
"<tbody>\n",
"\t<tr><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>⋯</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td></tr>\n",
"</tbody>\n",
"</table>\n"
],
"text/latex": [
"\\begin{tabular}{r|llllllllllllllllllllllllllllllllllllllllllllllllllllllllllllll}\n",
" Duration & Amount & InstallmentRatePercentage & ResidenceDuration & Age & NumberExistingCredits & NumberPeopleMaintenance & Telephone & ForeignWorker & Class & ⋯ & OtherInstallmentPlans.Bank & OtherInstallmentPlans.Stores & OtherInstallmentPlans.None & Housing.Rent & Housing.Own & Housing.ForFree & Job.UnemployedUnskilled & Job.UnskilledResident & Job.SkilledEmployee & Job.Management.SelfEmp.HighlyQualified\\\\\n",
"\\hline\n",
"\t 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & ⋯ & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0\\\\\n",
"\\end{tabular}\n"
],
"text/markdown": [
"\n",
"| Duration | Amount | InstallmentRatePercentage | ResidenceDuration | Age | NumberExistingCredits | NumberPeopleMaintenance | Telephone | ForeignWorker | Class | ⋯ | OtherInstallmentPlans.Bank | OtherInstallmentPlans.Stores | OtherInstallmentPlans.None | Housing.Rent | Housing.Own | Housing.ForFree | Job.UnemployedUnskilled | Job.UnskilledResident | Job.SkilledEmployee | Job.Management.SelfEmp.HighlyQualified |\n",
"|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|\n",
"| 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | ⋯ | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |\n",
"\n"
],
"text/plain": [
" Duration Amount InstallmentRatePercentage ResidenceDuration Age\n",
"1 0 0 0 0 0 \n",
" NumberExistingCredits NumberPeopleMaintenance Telephone ForeignWorker Class ⋯\n",
"1 0 0 0 0 0 ⋯\n",
" OtherInstallmentPlans.Bank OtherInstallmentPlans.Stores\n",
"1 0 0 \n",
" OtherInstallmentPlans.None Housing.Rent Housing.Own Housing.ForFree\n",
"1 0 0 0 0 \n",
" Job.UnemployedUnskilled Job.UnskilledResident Job.SkilledEmployee\n",
"1 0 0 0 \n",
" Job.Management.SelfEmp.HighlyQualified\n",
"1 0 "
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# Check on missing value\n",
"options(warn = -1) \n",
"german %>% summarise_all(funs(naratio))"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {},
"outputs": [],
"source": [
"# split dataset into 80/20 \n",
"set.seed(29)\n",
"id <- createDataPartition(y = german$Class, p = 0.8, list = FALSE)\n",
"\n",
"train_data <- german[id, ]\n",
"test_data <- german[-id, ]"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<table>\n",
"<thead><tr><th scope=col>Duration</th><th scope=col>Amount</th><th scope=col>InstallmentRatePercentage</th><th scope=col>ResidenceDuration</th><th scope=col>Age</th><th scope=col>NumberExistingCredits</th><th scope=col>NumberPeopleMaintenance</th><th scope=col>Telephone</th><th scope=col>ForeignWorker</th><th scope=col>Class</th><th scope=col>⋯</th><th scope=col>OtherInstallmentPlans.Bank</th><th scope=col>OtherInstallmentPlans.Stores</th><th scope=col>OtherInstallmentPlans.None</th><th scope=col>Housing.Rent</th><th scope=col>Housing.Own</th><th scope=col>Housing.ForFree</th><th scope=col>Job.UnemployedUnskilled</th><th scope=col>Job.UnskilledResident</th><th scope=col>Job.SkilledEmployee</th><th scope=col>Job.Management.SelfEmp.HighlyQualified</th></tr></thead>\n",
"<tbody>\n",
"\t<tr><td> 6 </td><td>1169</td><td>4 </td><td>4 </td><td>67 </td><td>2 </td><td>1 </td><td>0 </td><td>1 </td><td>Good</td><td>⋯ </td><td>0 </td><td>0 </td><td>1 </td><td>0 </td><td>1 </td><td>0 </td><td>0 </td><td>0 </td><td>1 </td><td>0 </td></tr>\n",
"\t<tr><td>48 </td><td>5951</td><td>2 </td><td>2 </td><td>22 </td><td>1 </td><td>1 </td><td>1 </td><td>1 </td><td>Bad </td><td>⋯ </td><td>0 </td><td>0 </td><td>1 </td><td>0 </td><td>1 </td><td>0 </td><td>0 </td><td>0 </td><td>1 </td><td>0 </td></tr>\n",
"\t<tr><td>12 </td><td>2096</td><td>2 </td><td>3 </td><td>49 </td><td>1 </td><td>2 </td><td>1 </td><td>1 </td><td>Good</td><td>⋯ </td><td>0 </td><td>0 </td><td>1 </td><td>0 </td><td>1 </td><td>0 </td><td>0 </td><td>1 </td><td>0 </td><td>0 </td></tr>\n",
"\t<tr><td>42 </td><td>7882</td><td>2 </td><td>4 </td><td>45 </td><td>1 </td><td>2 </td><td>1 </td><td>1 </td><td>Good</td><td>⋯ </td><td>0 </td><td>0 </td><td>1 </td><td>0 </td><td>0 </td><td>1 </td><td>0 </td><td>0 </td><td>1 </td><td>0 </td></tr>\n",
"\t<tr><td>24 </td><td>4870</td><td>3 </td><td>4 </td><td>53 </td><td>2 </td><td>2 </td><td>1 </td><td>1 </td><td>Bad </td><td>⋯ </td><td>0 </td><td>0 </td><td>1 </td><td>0 </td><td>0 </td><td>1 </td><td>0 </td><td>0 </td><td>1 </td><td>0 </td></tr>\n",
"</tbody>\n",
"</table>\n"
],
"text/latex": [
"\\begin{tabular}{r|llllllllllllllllllllllllllllllllllllllllllllllllllllllllllllll}\n",
" Duration & Amount & InstallmentRatePercentage & ResidenceDuration & Age & NumberExistingCredits & NumberPeopleMaintenance & Telephone & ForeignWorker & Class & ⋯ & OtherInstallmentPlans.Bank & OtherInstallmentPlans.Stores & OtherInstallmentPlans.None & Housing.Rent & Housing.Own & Housing.ForFree & Job.UnemployedUnskilled & Job.UnskilledResident & Job.SkilledEmployee & Job.Management.SelfEmp.HighlyQualified\\\\\n",
"\\hline\n",
"\t 6 & 1169 & 4 & 4 & 67 & 2 & 1 & 0 & 1 & Good & ⋯ & 0 & 0 & 1 & 0 & 1 & 0 & 0 & 0 & 1 & 0 \\\\\n",
"\t 48 & 5951 & 2 & 2 & 22 & 1 & 1 & 1 & 1 & Bad & ⋯ & 0 & 0 & 1 & 0 & 1 & 0 & 0 & 0 & 1 & 0 \\\\\n",
"\t 12 & 2096 & 2 & 3 & 49 & 1 & 2 & 1 & 1 & Good & ⋯ & 0 & 0 & 1 & 0 & 1 & 0 & 0 & 1 & 0 & 0 \\\\\n",
"\t 42 & 7882 & 2 & 4 & 45 & 1 & 2 & 1 & 1 & Good & ⋯ & 0 & 0 & 1 & 0 & 0 & 1 & 0 & 0 & 1 & 0 \\\\\n",
"\t 24 & 4870 & 3 & 4 & 53 & 2 & 2 & 1 & 1 & Bad & ⋯ & 0 & 0 & 1 & 0 & 0 & 1 & 0 & 0 & 1 & 0 \\\\\n",
"\\end{tabular}\n"
],
"text/markdown": [
"\n",
"| Duration | Amount | InstallmentRatePercentage | ResidenceDuration | Age | NumberExistingCredits | NumberPeopleMaintenance | Telephone | ForeignWorker | Class | ⋯ | OtherInstallmentPlans.Bank | OtherInstallmentPlans.Stores | OtherInstallmentPlans.None | Housing.Rent | Housing.Own | Housing.ForFree | Job.UnemployedUnskilled | Job.UnskilledResident | Job.SkilledEmployee | Job.Management.SelfEmp.HighlyQualified |\n",
"|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|\n",
"| 6 | 1169 | 4 | 4 | 67 | 2 | 1 | 0 | 1 | Good | ⋯ | 0 | 0 | 1 | 0 | 1 | 0 | 0 | 0 | 1 | 0 |\n",
"| 48 | 5951 | 2 | 2 | 22 | 1 | 1 | 1 | 1 | Bad | ⋯ | 0 | 0 | 1 | 0 | 1 | 0 | 0 | 0 | 1 | 0 |\n",
"| 12 | 2096 | 2 | 3 | 49 | 1 | 2 | 1 | 1 | Good | ⋯ | 0 | 0 | 1 | 0 | 1 | 0 | 0 | 1 | 0 | 0 |\n",
"| 42 | 7882 | 2 | 4 | 45 | 1 | 2 | 1 | 1 | Good | ⋯ | 0 | 0 | 1 | 0 | 0 | 1 | 0 | 0 | 1 | 0 |\n",
"| 24 | 4870 | 3 | 4 | 53 | 2 | 2 | 1 | 1 | Bad | ⋯ | 0 | 0 | 1 | 0 | 0 | 1 | 0 | 0 | 1 | 0 |\n",
"\n"
],
"text/plain": [
" Duration Amount InstallmentRatePercentage ResidenceDuration Age\n",
"1 6 1169 4 4 67 \n",
"2 48 5951 2 2 22 \n",
"3 12 2096 2 3 49 \n",
"4 42 7882 2 4 45 \n",
"5 24 4870 3 4 53 \n",
" NumberExistingCredits NumberPeopleMaintenance Telephone ForeignWorker Class ⋯\n",
"1 2 1 0 1 Good ⋯\n",
"2 1 1 1 1 Bad ⋯\n",
"3 1 2 1 1 Good ⋯\n",
"4 1 2 1 1 Good ⋯\n",
"5 2 2 1 1 Bad ⋯\n",
" OtherInstallmentPlans.Bank OtherInstallmentPlans.Stores\n",
"1 0 0 \n",
"2 0 0 \n",
"3 0 0 \n",
"4 0 0 \n",
"5 0 0 \n",
" OtherInstallmentPlans.None Housing.Rent Housing.Own Housing.ForFree\n",
"1 1 0 1 0 \n",
"2 1 0 1 0 \n",
"3 1 0 1 0 \n",
"4 1 0 0 1 \n",
"5 1 0 0 1 \n",
" Job.UnemployedUnskilled Job.UnskilledResident Job.SkilledEmployee\n",
"1 0 0 1 \n",
"2 0 0 1 \n",
"3 0 1 0 \n",
"4 0 0 1 \n",
"5 0 0 1 \n",
" Job.Management.SelfEmp.HighlyQualified\n",
"1 0 \n",
"2 0 \n",
"3 0 \n",
"4 0 \n",
"5 0 "
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"head(train_data,5)"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {},
"outputs": [],
"source": [
"#------------------------\n",
"# XGboost\n",
"#------------------------\n",
"# cross - validation condition setup\n",
"ctrl <- trainControl(method = \"repeatedcv\", number = 5, repeats = 6, summaryFunction = multiClassSummary)"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {},
"outputs": [],
"source": [
"#----------------\n",
"# XGBoost\n",
"#----------------\n",
"set.seed(29)\n",
"model_xgb_null2 <- train(Class ~ ., data = train_data, method = \"xgbTree\", preProcess = NULL, trControl = ctrl)"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"Confusion Matrix and Statistics\n",
"\n",
" Reference\n",
"Prediction Bad Good\n",
" Bad 28 18\n",
" Good 32 122\n",
" \n",
" Accuracy : 0.75 \n",
" 95% CI : (0.684, 0.8084)\n",
" No Information Rate : 0.7 \n",
" P-Value [Acc > NIR] : 0.06955 \n",
" \n",
" Kappa : 0.3622 \n",
" Mcnemar's Test P-Value : 0.06599 \n",
" \n",
" Sensitivity : 0.4667 \n",
" Specificity : 0.8714 \n",
" Pos Pred Value : 0.6087 \n",
" Neg Pred Value : 0.7922 \n",
" Prevalence : 0.3000 \n",
" Detection Rate : 0.1400 \n",
" Detection Prevalence : 0.2300 \n",
" Balanced Accuracy : 0.6690 \n",
" \n",
" 'Positive' Class : Bad \n",
" "
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# output: \n",
"pred1 <- predict(model_xgb_null2, test_data)\n",
"confusionMatrix(pred1, test_data$Class)"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {},
"outputs": [],
"source": [
"#------------------\n",
"# Logistic\n",
"#------------------\n",
"options(warn=-1)\n",
"model_logit_null2 <- train(Class ~ ., data = train_data, method = \"glm\", \n",
" family = \"binomial\", preProcess = NULL, trControl = ctrl)"
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"Confusion Matrix and Statistics\n",
"\n",
" Reference\n",
"Prediction Bad Good\n",
" Bad 32 19\n",
" Good 28 121\n",
" \n",
" Accuracy : 0.765 \n",
" 95% CI : (0.7, 0.8219)\n",
" No Information Rate : 0.7 \n",
" P-Value [Acc > NIR] : 0.02493 \n",
" \n",
" Kappa : 0.4154 \n",
" Mcnemar's Test P-Value : 0.24324 \n",
" \n",
" Sensitivity : 0.5333 \n",
" Specificity : 0.8643 \n",
" Pos Pred Value : 0.6275 \n",
" Neg Pred Value : 0.8121 \n",
" Prevalence : 0.3000 \n",
" Detection Rate : 0.1600 \n",
" Detection Prevalence : 0.2550 \n",
" Balanced Accuracy : 0.6988 \n",
" \n",
" 'Positive' Class : Bad \n",
" "
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# output of Logistic: \n",
"options(warn=-1)\n",
"pred2 <- predict(model_logit_null2, test_data)\n",
"confusionMatrix(pred2, test_data$Class)"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {},
"outputs": [],
"source": [
"#------------------\n",
"# KNN\n",
"#------------------\n",
"options(warn=-1)\n",
"model_KNN_null2 <- train(Class ~ ., data = train_data, method = \"knn\", preProcess = NULL, trControl = ctrl)"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"Confusion Matrix and Statistics\n",
"\n",
" Reference\n",
"Prediction Bad Good\n",
" Bad 11 15\n",
" Good 49 125\n",
" \n",
" Accuracy : 0.68 \n",
" 95% CI : (0.6105, 0.744)\n",
" No Information Rate : 0.7 \n",
" P-Value [Acc > NIR] : 0.7579 \n",
" \n",
" Kappa : 0.0909 \n",
" Mcnemar's Test P-Value : 3.707e-05 \n",
" \n",
" Sensitivity : 0.1833 \n",
" Specificity : 0.8929 \n",
" Pos Pred Value : 0.4231 \n",
" Neg Pred Value : 0.7184 \n",
" Prevalence : 0.3000 \n",
" Detection Rate : 0.0550 \n",
" Detection Prevalence : 0.1300 \n",
" Balanced Accuracy : 0.5381 \n",
" \n",
" 'Positive' Class : Bad \n",
" "
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# output of KNN: \n",
"options(warn=-1)\n",
"predKNN <- predict(model_KNN_null2, test_data)\n",
"confusionMatrix(predKNN, test_data$Class)"
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {},
"outputs": [],
"source": [
"#------------------\n",
"# Random Forest\n",
"#------------------\n",
"options(warn=-1)\n",
"model_RF_null2 <- train(Class ~ ., data = train_data, method = \"rf\", preProcess = NULL, trControl = ctrl)"
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"Confusion Matrix and Statistics\n",
"\n",
" Reference\n",
"Prediction Bad Good\n",
" Bad 30 15\n",
" Good 30 125\n",
" \n",
" Accuracy : 0.775 \n",
" 95% CI : (0.7108, 0.8309)\n",
" No Information Rate : 0.7 \n",
" P-Value [Acc > NIR] : 0.01113 \n",
" \n",
" Kappa : 0.4231 \n",
" Mcnemar's Test P-Value : 0.03689 \n",
" \n",
" Sensitivity : 0.5000 \n",
" Specificity : 0.8929 \n",
" Pos Pred Value : 0.6667 \n",
" Neg Pred Value : 0.8065 \n",
" Prevalence : 0.3000 \n",
" Detection Rate : 0.1500 \n",
" Detection Prevalence : 0.2250 \n",
" Balanced Accuracy : 0.6964 \n",
" \n",
" 'Positive' Class : Bad \n",
" "
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# output of Random Forest: \n",
"options(warn=-1)\n",
"predRF <- predict(model_RF_null2, test_data)\n",
"confusionMatrix(predRF, test_data$Class)"
]
},
{
"cell_type": "code",
"execution_count": 17,
"metadata": {},
"outputs": [],
"source": [
"#------------------\n",
"# Support Vector Machine\n",
"#------------------\n",
"options(warn=-1)\n",
"model_SVM_null2 <- train(Class ~ ., data = train_data, method = \"svmLinear\", preProcess = NULL, trControl = ctrl)"
]
},
{
"cell_type": "code",
"execution_count": 18,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"Confusion Matrix and Statistics\n",
"\n",
" Reference\n",
"Prediction Bad Good\n",
" Bad 16 7\n",
" Good 44 133\n",
" \n",
" Accuracy : 0.745 \n",
" 95% CI : (0.6787, 0.8039)\n",
" No Information Rate : 0.7 \n",
" P-Value [Acc > NIR] : 0.09344 \n",
" \n",
" Kappa : 0.263 \n",
" Mcnemar's Test P-Value : 4.631e-07 \n",
" \n",
" Sensitivity : 0.2667 \n",
" Specificity : 0.9500 \n",
" Pos Pred Value : 0.6957 \n",
" Neg Pred Value : 0.7514 \n",
" Prevalence : 0.3000 \n",
" Detection Rate : 0.0800 \n",
" Detection Prevalence : 0.1150 \n",
" Balanced Accuracy : 0.6083 \n",
" \n",
" 'Positive' Class : Bad \n",
" "
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# output of SVM: \n",
"options(warn=-1)\n",
"predSVM <- predict(model_SVM_null2, test_data)\n",
"confusionMatrix(predSVM, test_data$Class)"
]
},
{
"cell_type": "code",
"execution_count": 19,
"metadata": {},
"outputs": [],
"source": [
"#plotting results\n",
"my_box_plot <- function(model1, model2, model3, model4, model5) {\n",
" \n",
" u1 <- model1$resample %>% \n",
" select(-Resample) %>% \n",
" mutate(Model = \"XGBoost\")\n",
" \n",
" u2 <- model2$resample %>% \n",
" select(-Resample) %>% \n",
" mutate(Model = \"Logistic\")\n",
" \n",
" u3 <- model3$resample %>% \n",
" select(-Resample) %>% \n",
" mutate(Model = \"KNN\")\n",
" \n",
" u4 <- model4$resample %>% \n",
" select(-Resample) %>% \n",
" mutate(Model = \"Random Forest\") \n",
" \n",
" u5 <- model5$resample %>% \n",
" select(-Resample) %>% \n",
" mutate(Model = \"SVM\") \n",
" \n",
" all_results <- rbind(u1, u2, u3, u4, u5)\n",
" \n",
" plot <- all_results_4 <- all_results %>% \n",
" select(Accuracy, Pos_Pred_Value, Neg_Pred_Value, Kappa, Model) %>% \n",
" gather(Criteria, Value, -Model) %>% \n",
" ggplot(aes(Model, Value, color = Model, fill = Model)) + \n",
" geom_boxplot(alpha = 0.3, show.legend = FALSE) + \n",
" facet_wrap(~ Criteria, scales = \"free\") + \n",
" coord_flip() + \n",
" theme_minimal() + \n",
" labs(x = NULL, y = NULL, \n",
" title = \"A comparison between two models based on four Criterion\")\n",
" \n",
" return(plot) \n",
"}"
]
},
{
"cell_type": "code",
"execution_count": 20,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAABLAAAAPACAYAAAAlpICAAAAEGWlDQ1BrQ0dDb2xvclNwYWNl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\nBBBAAAEEEEAAAQQQQAABBBBAILPA/wP6n3fwuE3o2gAAAABJRU5ErkJggg==",
"text/plain": [
"plot without title"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"options(repr.plot.width = 10, repr.plot.height = 8)\n",
"my_box_plot(model1 = model_xgb_null2, model2 = model_logit_null2, model3 = model_KNN_null2, \n",
" model4 = model_RF_null2, model5 = model_SVM_null2) + \n",
" labs(subtitle = \"Data: german_credit.csv\")"
]
},
{
"cell_type": "code",
"execution_count": 21,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"\n",
"Attaching package: ‘caretEnsemble’\n",
"\n",
"The following object is masked from ‘package:ggplot2’:\n",
"\n",
" autoplot\n",
"\n"
]
}
],
"source": [
"library(caretEnsemble)"
]
},
{
"cell_type": "code",
"execution_count": 22,
"metadata": {},
"outputs": [],
"source": [
"set.seed(1)\n",
"myControl <- trainControl(method=\"cv\", number=5, summaryFunction = multiClassSummary)\n",
"model_list1 <- caretList(Class ~., data = train_data, trControl = myControl, metric = 'Accuracy', \n",
" methodList = c(\"glm\", \"rf\", \"svmLinear\", \"knn\", \"xgbTree\")\n",
" )"
]
},
{
"cell_type": "code",
"execution_count": 23,
"metadata": {},
"outputs": [],
"source": [
"# gathering result\n",
"total_df <- bind_rows(model_list1$glm$resample %>% \n",
" select(-Resample) %>% \n",
" mutate(Model = \"Logistic\"), \n",
" \n",
" model_list1$rf$resample %>% \n",
" select(-Resample) %>% \n",
" mutate(Model = \"RF\"), \n",
" \n",
" model_list1$knn$resample %>% \n",
" select(-Resample) %>% \n",
" mutate(Model = \"KNN\"), \n",
" \n",
" model_list1$svmLinear$resample %>% \n",
" select(-Resample) %>% \n",
" mutate(Model = \"SVM\"), \n",
" \n",
" model_list1$xgbTree$resample %>% \n",
" select(-Resample) %>% \n",
" mutate(Model = \"XgbTree\"),\n",
" )"
]
},
{
"cell_type": "code",
"execution_count": 24,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"$glm\n",
"Generalized Linear Model \n",
"\n",
"800 samples\n",
" 61 predictor\n",
" 2 classes: 'Bad', 'Good' \n",
"\n",
"No pre-processing\n",
"Resampling: Cross-Validated (5 fold) \n",
"Summary of sample sizes: 640, 640, 640, 640, 640 \n",
"Resampling results:\n",
"\n",
" Accuracy Kappa F1 Sensitivity Specificity Pos_Pred_Value\n",
" 0.74125 0.3442956 0.5174571 0.4625 0.8607143 0.5892711 \n",
" Neg_Pred_Value Precision Recall Detection_Rate Balanced_Accuracy\n",
" 0.7889118 0.5892711 0.4625 0.13875 0.6616071 \n",
"\n",
"\n",
"$rf\n",
"Random Forest \n",
"\n",
"800 samples\n",
" 61 predictor\n",
" 2 classes: 'Bad', 'Good' \n",
"\n",
"No pre-processing\n",
"Resampling: Cross-Validated (5 fold) \n",
"Summary of sample sizes: 640, 640, 640, 640, 640 \n",
"Resampling results across tuning parameters:\n",
"\n",
" mtry Accuracy Kappa F1 Sensitivity Specificity\n",
" 2 0.71625 0.08292286 0.1224083 0.06666667 0.9946429 \n",
" 31 0.74500 0.33617769 0.4995445 0.42500000 0.8821429 \n",
" 61 0.73625 0.32466915 0.4984686 0.43750000 0.8642857 \n",
" Pos_Pred_Value Neg_Pred_Value Precision Recall Detection_Rate\n",
" 0.8833333 0.7132248 0.8833333 0.06666667 0.02000 \n",
" 0.6127260 0.7817733 0.6127260 0.42500000 0.12750 \n",
" 0.5828571 0.7820083 0.5828571 0.43750000 0.13125 \n",
" Balanced_Accuracy\n",
" 0.5306548 \n",
" 0.6535714 \n",
" 0.6508929 \n",
"\n",
"Accuracy was used to select the optimal model using the largest value.\n",
"The final value used for the model was mtry = 31.\n",
"\n",
"$svmLinear\n",
"Support Vector Machines with Linear Kernel \n",
"\n",
"800 samples\n",
" 61 predictor\n",
" 2 classes: 'Bad', 'Good' \n",
"\n",
"No pre-processing\n",
"Resampling: Cross-Validated (5 fold) \n",
"Summary of sample sizes: 640, 640, 640, 640, 640 \n",
"Resampling results:\n",
"\n",
" Accuracy Kappa F1 Sensitivity Specificity Pos_Pred_Value\n",
" 0.70875 0.2061642 0.3754318 0.2958333 0.8857143 0.521303 \n",
" Neg_Pred_Value Precision Recall Detection_Rate Balanced_Accuracy\n",
" 0.746241 0.521303 0.2958333 0.08875 0.5907738 \n",
"\n",
"Tuning parameter 'C' was held constant at a value of 1\n",
"\n",
"$knn\n",
"k-Nearest Neighbors \n",
"\n",
"800 samples\n",
" 61 predictor\n",
" 2 classes: 'Bad', 'Good' \n",
"\n",
"No pre-processing\n",
"Resampling: Cross-Validated (5 fold) \n",
"Summary of sample sizes: 640, 640, 640, 640, 640 \n",
"Resampling results across tuning parameters:\n",
"\n",
" k Accuracy Kappa F1 Sensitivity Specificity Pos_Pred_Value\n",
" 5 0.65500 0.07788497 0.2887805 0.2333333 0.8357143 0.3803050 \n",
" 7 0.66375 0.08940503 0.2871515 0.2250000 0.8517857 0.4073436 \n",
" 9 0.66875 0.06179484 0.2327249 0.1666667 0.8839286 0.3947368 \n",
" Neg_Pred_Value Precision Recall Detection_Rate Balanced_Accuracy\n",
" 0.7177045 0.3803050 0.2333333 0.0700 0.5345238 \n",
" 0.7190070 0.4073436 0.2250000 0.0675 0.5383929 \n",
" 0.7119150 0.3947368 0.1666667 0.0500 0.5252976 \n",
"\n",
"Accuracy was used to select the optimal model using the largest value.\n",
"The final value used for the model was k = 9.\n",
"\n",
"$xgbTree\n",
"eXtreme Gradient Boosting \n",
"\n",
"800 samples\n",
" 61 predictor\n",
" 2 classes: 'Bad', 'Good' \n",
"\n",
"No pre-processing\n",
"Resampling: Cross-Validated (5 fold) \n",
"Summary of sample sizes: 640, 640, 640, 640, 640 \n",
"Resampling results across tuning parameters:\n",
"\n",
" eta max_depth colsample_bytree subsample nrounds Accuracy Kappa \n",
" 0.3 1 0.6 0.50 50 0.74875 0.3490313\n",
" 0.3 1 0.6 0.50 100 0.74375 0.3473530\n",
" 0.3 1 0.6 0.50 150 0.73625 0.3329469\n",
" 0.3 1 0.6 0.75 50 0.74875 0.3446397\n",
" 0.3 1 0.6 0.75 100 0.75125 0.3627501\n",
" 0.3 1 0.6 0.75 150 0.75250 0.3654599\n",
" 0.3 1 0.6 1.00 50 0.74250 0.3101304\n",
" 0.3 1 0.6 1.00 100 0.75375 0.3537581\n",
" 0.3 1 0.6 1.00 150 0.75375 0.3629174\n",
" 0.3 1 0.8 0.50 50 0.75125 0.3638801\n",
" 0.3 1 0.8 0.50 100 0.74375 0.3531656\n",
" 0.3 1 0.8 0.50 150 0.75250 0.3846429\n",
" 0.3 1 0.8 0.75 50 0.74500 0.3322427\n",
" 0.3 1 0.8 0.75 100 0.75000 0.3601518\n",
" 0.3 1 0.8 0.75 150 0.75125 0.3654478\n",
" 0.3 1 0.8 1.00 50 0.74000 0.2987244\n",
" 0.3 1 0.8 1.00 100 0.75500 0.3603519\n",
" 0.3 1 0.8 1.00 150 0.75500 0.3652824\n",
" 0.3 2 0.6 0.50 50 0.74250 0.3494121\n",
" 0.3 2 0.6 0.50 100 0.74000 0.3580664\n",
" 0.3 2 0.6 0.50 150 0.74000 0.3453370\n",
" 0.3 2 0.6 0.75 50 0.74750 0.3501097\n",
" 0.3 2 0.6 0.75 100 0.74375 0.3471118\n",
" 0.3 2 0.6 0.75 150 0.74750 0.3625361\n",
" 0.3 2 0.6 1.00 50 0.75125 0.3573732\n",
" 0.3 2 0.6 1.00 100 0.75125 0.3734646\n",
" 0.3 2 0.6 1.00 150 0.74375 0.3583347\n",
" 0.3 2 0.8 0.50 50 0.73625 0.3267442\n",
" 0.3 2 0.8 0.50 100 0.74000 0.3526868\n",
" 0.3 2 0.8 0.50 150 0.72875 0.3333210\n",
" 0.3 2 0.8 0.75 50 0.73625 0.3272854\n",
" 0.3 2 0.8 0.75 100 0.74000 0.3479546\n",
" 0.3 2 0.8 0.75 150 0.72500 0.3162383\n",
" 0.3 2 0.8 1.00 50 0.75250 0.3618999\n",
" 0.3 2 0.8 1.00 100 0.74625 0.3568292\n",
" 0.3 2 0.8 1.00 150 0.74625 0.3633561\n",
" 0.3 3 0.6 0.50 50 0.74125 0.3652028\n",
" 0.3 3 0.6 0.50 100 0.72000 0.3135578\n",
" 0.3 3 0.6 0.50 150 0.72500 0.3144093\n",
" 0.3 3 0.6 0.75 50 0.72750 0.3178622\n",
" 0.3 3 0.6 0.75 100 0.73875 0.3556922\n",
" 0.3 3 0.6 0.75 150 0.73500 0.3426107\n",
" 0.3 3 0.6 1.00 50 0.74625 0.3530494\n",
" 0.3 3 0.6 1.00 100 0.75250 0.3827160\n",
" 0.3 3 0.6 1.00 150 0.73000 0.3281793\n",
" 0.3 3 0.8 0.50 50 0.72750 0.3187668\n",
" 0.3 3 0.8 0.50 100 0.72375 0.3120706\n",
" 0.3 3 0.8 0.50 150 0.72000 0.3030089\n",
" 0.3 3 0.8 0.75 50 0.73250 0.3331320\n",
" 0.3 3 0.8 0.75 100 0.73125 0.3430894\n",
" 0.3 3 0.8 0.75 150 0.73250 0.3475877\n",
" 0.3 3 0.8 1.00 50 0.74875 0.3549551\n",
" 0.3 3 0.8 1.00 100 0.74500 0.3534659\n",
" 0.3 3 0.8 1.00 150 0.74375 0.3589559\n",
" 0.4 1 0.6 0.50 50 0.71875 0.2630144\n",
" 0.4 1 0.6 0.50 100 0.73625 0.3325299\n",
" 0.4 1 0.6 0.50 150 0.72375 0.3101628\n",
" 0.4 1 0.6 0.75 50 0.74250 0.3304403\n",
" 0.4 1 0.6 0.75 100 0.73750 0.3367111\n",
" 0.4 1 0.6 0.75 150 0.73750 0.3372079\n",
" 0.4 1 0.6 1.00 50 0.74875 0.3375849\n",
" 0.4 1 0.6 1.00 100 0.75250 0.3552099\n",
" 0.4 1 0.6 1.00 150 0.75125 0.3598720\n",
" 0.4 1 0.8 0.50 50 0.75125 0.3552716\n",
" 0.4 1 0.8 0.50 100 0.74125 0.3503843\n",
" 0.4 1 0.8 0.50 150 0.73750 0.3486726\n",
" 0.4 1 0.8 0.75 50 0.74500 0.3366851\n",
" 0.4 1 0.8 0.75 100 0.74125 0.3468829\n",
" 0.4 1 0.8 0.75 150 0.74000 0.3443909\n",
" 0.4 1 0.8 1.00 50 0.75125 0.3477909\n",
" 0.4 1 0.8 1.00 100 0.75500 0.3639852\n",
" 0.4 1 0.8 1.00 150 0.75500 0.3667444\n",
" 0.4 2 0.6 0.50 50 0.72250 0.3108055\n",
" 0.4 2 0.6 0.50 100 0.73375 0.3218735\n",
" 0.4 2 0.6 0.50 150 0.72000 0.2991138\n",
" 0.4 2 0.6 0.75 50 0.74250 0.3511514\n",
" 0.4 2 0.6 0.75 100 0.71750 0.2989765\n",
" 0.4 2 0.6 0.75 150 0.73000 0.3363149\n",
" 0.4 2 0.6 1.00 50 0.74625 0.3529688\n",
" 0.4 2 0.6 1.00 100 0.74625 0.3696205\n",
" 0.4 2 0.6 1.00 150 0.73625 0.3489767\n",
" 0.4 2 0.8 0.50 50 0.71750 0.2973586\n",
" 0.4 2 0.8 0.50 100 0.72250 0.3126319\n",
" 0.4 2 0.8 0.50 150 0.72250 0.3175927\n",
" 0.4 2 0.8 0.75 50 0.74625 0.3633263\n",
" 0.4 2 0.8 0.75 100 0.74875 0.3759947\n",
" 0.4 2 0.8 0.75 150 0.75000 0.3798598\n",
" 0.4 2 0.8 1.00 50 0.75125 0.3704778\n",
" 0.4 2 0.8 1.00 100 0.74625 0.3669207\n",
" 0.4 2 0.8 1.00 150 0.73625 0.3524856\n",
" 0.4 3 0.6 0.50 50 0.71750 0.3184480\n",
" 0.4 3 0.6 0.50 100 0.70000 0.2703103\n",
" 0.4 3 0.6 0.50 150 0.70875 0.2953307\n",
" 0.4 3 0.6 0.75 50 0.73875 0.3500922\n",
" 0.4 3 0.6 0.75 100 0.73125 0.3338355\n",
" 0.4 3 0.6 0.75 150 0.72000 0.3097169\n",
" 0.4 3 0.6 1.00 50 0.74000 0.3528036\n",
" 0.4 3 0.6 1.00 100 0.74500 0.3654121\n",
" 0.4 3 0.6 1.00 150 0.74250 0.3610017\n",
" 0.4 3 0.8 0.50 50 0.71125 0.2987439\n",
" 0.4 3 0.8 0.50 100 0.72000 0.3156129\n",
" 0.4 3 0.8 0.50 150 0.72875 0.3407120\n",
" 0.4 3 0.8 0.75 50 0.71875 0.2907703\n",
" 0.4 3 0.8 0.75 100 0.71250 0.2928313\n",
" 0.4 3 0.8 0.75 150 0.73375 0.3452324\n",
" 0.4 3 0.8 1.00 50 0.72500 0.3094106\n",
" 0.4 3 0.8 1.00 100 0.72250 0.3060964\n",
" 0.4 3 0.8 1.00 150 0.72625 0.3095661\n",
" F1 Sensitivity Specificity Pos_Pred_Value Neg_Pred_Value\n",
" 0.5115670 0.4416667 0.8803571 0.6122587 0.7867367 \n",
" 0.5174361 0.4625000 0.8642857 0.5919237 0.7903084 \n",
" 0.5100357 0.4583333 0.8553571 0.5761326 0.7866562 \n",
" 0.5056373 0.4291667 0.8857143 0.6173251 0.7836889 \n",
" 0.5267988 0.4625000 0.8750000 0.6135056 0.7917404 \n",
" 0.5284645 0.4625000 0.8767857 0.6184458 0.7919871 \n",
" 0.4664225 0.3791667 0.8982143 0.6116845 0.7719348 \n",
" 0.5095059 0.4291667 0.8928571 0.6325343 0.7853022 \n",
" 0.5225947 0.4500000 0.8839286 0.6280088 0.7896356 \n",
" 0.5281141 0.4666667 0.8732143 0.6118805 0.7929556 \n",
" 0.5255489 0.4750000 0.8589286 0.5902841 0.7927203 \n",
" 0.5540031 0.5125000 0.8553571 0.6066420 0.8037899 \n",
" 0.4942610 0.4166667 0.8857143 0.6124339 0.7800674 \n",
" 0.5253216 0.4625000 0.8732143 0.6087434 0.7914440 \n",
" 0.5303796 0.4708333 0.8714286 0.6098020 0.7938857 \n",
" 0.4544884 0.3625000 0.9017857 0.6121653 0.7676442 \n",
" 0.5169938 0.4375000 0.8910714 0.6354230 0.7871598 \n",
" 0.5237675 0.4500000 0.8857143 0.6316072 0.7899924 \n",
" 0.5217416 0.4708333 0.8589286 0.5924126 0.7915908 \n",
" 0.5371690 0.5041667 0.8410714 0.5790936 0.7985897 \n",
" 0.5204518 0.4708333 0.8553571 0.5865443 0.7905596 \n",
" 0.5149239 0.4500000 0.8750000 0.6095065 0.7883361 \n",
" 0.5171721 0.4583333 0.8660714 0.5970107 0.7887842 \n",
" 0.5321764 0.4791667 0.8625000 0.6023941 0.7945307 \n",
" 0.5189346 0.4500000 0.8803571 0.6182068 0.7892090 \n",
" 0.5409316 0.4916667 0.8625000 0.6056650 0.7988111 \n",
" 0.5323096 0.4875000 0.8535714 0.5891342 0.7955724 \n",
" 0.5013845 0.4458333 0.8607143 0.5758253 0.7842997 \n",
" 0.5303566 0.4916667 0.8464286 0.5790563 0.7957129 \n",
" 0.5214064 0.4916667 0.8303571 0.5572018 0.7920718 \n",
" 0.5021655 0.4458333 0.8607143 0.5775244 0.7840866 \n",
" 0.5238554 0.4791667 0.8517857 0.5830342 0.7927877 \n",
" 0.5041073 0.4708333 0.8339286 0.5477915 0.7870372 \n",
" 0.5234202 0.4541667 0.8803571 0.6204280 0.7902281 \n",
" 0.5266046 0.4708333 0.8642857 0.5997318 0.7922281 \n",
" 0.5353874 0.4875000 0.8571429 0.5956938 0.7960802 \n",
" 0.5448143 0.5166667 0.8375000 0.5795091 0.8018638 \n",
" 0.5080323 0.4833333 0.8214286 0.5385534 0.7879367 \n",
" 0.5023485 0.4625000 0.8375000 0.5508389 0.7842704 \n",
" 0.5026074 0.4625000 0.8410714 0.5563564 0.7857247 \n",
" 0.5351436 0.5041667 0.8392857 0.5806956 0.7987719 \n",
" 0.5234366 0.4916667 0.8392857 0.5700975 0.7952651 \n",
" 0.5211464 0.4625000 0.8678571 0.6017064 0.7905890 \n",
" 0.5514951 0.5083333 0.8571429 0.6062442 0.8029201 \n",
" 0.5127103 0.4750000 0.8392857 0.5605237 0.7888407 \n",
" 0.5038466 0.4625000 0.8410714 0.5582251 0.7852434 \n",
" 0.5009336 0.4625000 0.8357143 0.5485205 0.7839957 \n",
" 0.4942653 0.4583333 0.8321429 0.5400190 0.7822607 \n",
" 0.5155717 0.4750000 0.8428571 0.5677413 0.7894366 \n",
" 0.5303403 0.5083333 0.8267857 0.5573572 0.7973748 \n",
" 0.5337610 0.5125000 0.8267857 0.5640092 0.7987947 \n",
" 0.5198443 0.4541667 0.8750000 0.6118627 0.7891993 \n",
" 0.5233406 0.4708333 0.8625000 0.5963848 0.7924553 \n",
" 0.5325392 0.4916667 0.8517857 0.5872019 0.7970646 \n",
" 0.4414996 0.3708333 0.8678571 0.5495621 0.7629792 \n",
" 0.5093641 0.4583333 0.8553571 0.5751888 0.7868335 \n",
" 0.4985491 0.4583333 0.8375000 0.5479016 0.7830793 \n",
" 0.4963104 0.4250000 0.8785714 0.5982760 0.7812300 \n",
" 0.5130979 0.4625000 0.8553571 0.5778397 0.7880471 \n",
" 0.5138779 0.4625000 0.8553571 0.5790372 0.7878433 \n",
" 0.4953803 0.4125000 0.8928571 0.6237169 0.7802237 \n",
" 0.5139447 0.4375000 0.8875000 0.6265747 0.7865981 \n",
" 0.5226322 0.4541667 0.8785714 0.6187487 0.7898073 \n",
" 0.5159818 0.4458333 0.8821429 0.6175835 0.7883714 \n",
" 0.5251297 0.4791667 0.8535714 0.5868027 0.7930996 \n",
" 0.5288359 0.4916667 0.8428571 0.5746516 0.7947728 \n",
" 0.5006911 0.4291667 0.8803571 0.6042179 0.7829462 \n",
" 0.5208334 0.4708333 0.8571429 0.5858226 0.7911151 \n",
" 0.5190225 0.4708333 0.8553571 0.5845809 0.7909952 \n",
" 0.5057335 0.4250000 0.8910714 0.6265675 0.7834732 \n",
" 0.5219343 0.4458333 0.8875000 0.6348284 0.7889602 \n",
" 0.5259774 0.4541667 0.8839286 0.6280692 0.7909199 \n",
" 0.5004428 0.4625000 0.8339286 0.5512584 0.7834983 \n",
" 0.4988998 0.4416667 0.8589286 0.5748221 0.7821074 \n",
" 0.4890244 0.4500000 0.8357143 0.5399437 0.7805534 \n",
" 0.5244432 0.4750000 0.8571429 0.5898740 0.7923976 \n",
" 0.4930256 0.4583333 0.8285714 0.5347326 0.7812430 \n",
" 0.5237200 0.4958333 0.8303571 0.5558137 0.7936975 \n",
" 0.5209934 0.4625000 0.8678571 0.6019034 0.7906461 \n",
" 0.5433037 0.5041667 0.8500000 0.5934254 0.8002709 \n",
" 0.5305998 0.4958333 0.8392857 0.5755710 0.7952004 \n",
" 0.4899238 0.4583333 0.8285714 0.5338090 0.7822306 \n",
" 0.5030482 0.4708333 0.8303571 0.5446951 0.7858908 \n",
" 0.5088244 0.4833333 0.8250000 0.5453628 0.7890582 \n",
" 0.5343728 0.4875000 0.8571429 0.6003662 0.7964669 \n",
" 0.5483081 0.5083333 0.8517857 0.5972154 0.8017453 \n",
" 0.5515750 0.5125000 0.8517857 0.5990483 0.8030871 \n",
" 0.5367382 0.4833333 0.8660714 0.6096981 0.7968812 \n",
" 0.5395594 0.5000000 0.8517857 0.5918111 0.7997823 \n",
" 0.5354683 0.5083333 0.8339286 0.5685198 0.7986020 \n",
" 0.5176804 0.5041667 0.8089286 0.5343391 0.7918032 \n",
" 0.4803543 0.4625000 0.8017857 0.5019251 0.7768809 \n",
" 0.5001478 0.4875000 0.8035714 0.5162603 0.7856797 \n",
" 0.5287867 0.4916667 0.8446429 0.5755051 0.7954919 \n",
" 0.5180500 0.4833333 0.8375000 0.5634594 0.7912679 \n",
" 0.5030619 0.4750000 0.8250000 0.5381508 0.7862377 \n",
" 0.5291169 0.4875000 0.8482143 0.5928798 0.7946843 \n",
" 0.5385574 0.5000000 0.8500000 0.5959736 0.7993301 \n",
" 0.5362988 0.5000000 0.8464286 0.5913461 0.7986259 \n",
" 0.5007228 0.4875000 0.8071429 0.5193875 0.7869924 \n",
" 0.5101498 0.4916667 0.8178571 0.5359917 0.7906531 \n",
" 0.5296429 0.5166667 0.8196429 0.5522947 0.7996831 \n",
" 0.4799966 0.4333333 0.8410714 0.5410904 0.7760501 \n",
" 0.4915119 0.4625000 0.8196429 0.5297298 0.7805596 \n",
" 0.5286937 0.4958333 0.8357143 0.5775526 0.7945973 \n",
" 0.4952404 0.4500000 0.8428571 0.5555856 0.7816187 \n",
" 0.4944437 0.4541667 0.8375000 0.5485550 0.7820533 \n",
" 0.4929166 0.4500000 0.8446429 0.5541131 0.7829103 \n",
" Precision Recall Detection_Rate Balanced_Accuracy\n",
" 0.6122587 0.4416667 0.13250 0.6610119 \n",
" 0.5919237 0.4625000 0.13875 0.6633929 \n",
" 0.5761326 0.4583333 0.13750 0.6568452 \n",
" 0.6173251 0.4291667 0.12875 0.6574405 \n",
" 0.6135056 0.4625000 0.13875 0.6687500 \n",
" 0.6184458 0.4625000 0.13875 0.6696429 \n",
" 0.6116845 0.3791667 0.11375 0.6386905 \n",
" 0.6325343 0.4291667 0.12875 0.6610119 \n",
" 0.6280088 0.4500000 0.13500 0.6669643 \n",
" 0.6118805 0.4666667 0.14000 0.6699405 \n",
" 0.5902841 0.4750000 0.14250 0.6669643 \n",
" 0.6066420 0.5125000 0.15375 0.6839286 \n",
" 0.6124339 0.4166667 0.12500 0.6511905 \n",
" 0.6087434 0.4625000 0.13875 0.6678571 \n",
" 0.6098020 0.4708333 0.14125 0.6711310 \n",
" 0.6121653 0.3625000 0.10875 0.6321429 \n",
" 0.6354230 0.4375000 0.13125 0.6642857 \n",
" 0.6316072 0.4500000 0.13500 0.6678571 \n",
" 0.5924126 0.4708333 0.14125 0.6648810 \n",
" 0.5790936 0.5041667 0.15125 0.6726190 \n",
" 0.5865443 0.4708333 0.14125 0.6630952 \n",
" 0.6095065 0.4500000 0.13500 0.6625000 \n",
" 0.5970107 0.4583333 0.13750 0.6622024 \n",
" 0.6023941 0.4791667 0.14375 0.6708333 \n",
" 0.6182068 0.4500000 0.13500 0.6651786 \n",
" 0.6056650 0.4916667 0.14750 0.6770833 \n",
" 0.5891342 0.4875000 0.14625 0.6705357 \n",
" 0.5758253 0.4458333 0.13375 0.6532738 \n",
" 0.5790563 0.4916667 0.14750 0.6690476 \n",
" 0.5572018 0.4916667 0.14750 0.6610119 \n",
" 0.5775244 0.4458333 0.13375 0.6532738 \n",
" 0.5830342 0.4791667 0.14375 0.6654762 \n",
" 0.5477915 0.4708333 0.14125 0.6523810 \n",
" 0.6204280 0.4541667 0.13625 0.6672619 \n",
" 0.5997318 0.4708333 0.14125 0.6675595 \n",
" 0.5956938 0.4875000 0.14625 0.6723214 \n",
" 0.5795091 0.5166667 0.15500 0.6770833 \n",
" 0.5385534 0.4833333 0.14500 0.6523810 \n",
" 0.5508389 0.4625000 0.13875 0.6500000 \n",
" 0.5563564 0.4625000 0.13875 0.6517857 \n",
" 0.5806956 0.5041667 0.15125 0.6717262 \n",
" 0.5700975 0.4916667 0.14750 0.6654762 \n",
" 0.6017064 0.4625000 0.13875 0.6651786 \n",
" 0.6062442 0.5083333 0.15250 0.6827381 \n",
" 0.5605237 0.4750000 0.14250 0.6571429 \n",
" 0.5582251 0.4625000 0.13875 0.6517857 \n",
" 0.5485205 0.4625000 0.13875 0.6491071 \n",
" 0.5400190 0.4583333 0.13750 0.6452381 \n",
" 0.5677413 0.4750000 0.14250 0.6589286 \n",
" 0.5573572 0.5083333 0.15250 0.6675595 \n",
" 0.5640092 0.5125000 0.15375 0.6696429 \n",
" 0.6118627 0.4541667 0.13625 0.6645833 \n",
" 0.5963848 0.4708333 0.14125 0.6666667 \n",
" 0.5872019 0.4916667 0.14750 0.6717262 \n",
" 0.5495621 0.3708333 0.11125 0.6193452 \n",
" 0.5751888 0.4583333 0.13750 0.6568452 \n",
" 0.5479016 0.4583333 0.13750 0.6479167 \n",
" 0.5982760 0.4250000 0.12750 0.6517857 \n",
" 0.5778397 0.4625000 0.13875 0.6589286 \n",
" 0.5790372 0.4625000 0.13875 0.6589286 \n",
" 0.6237169 0.4125000 0.12375 0.6526786 \n",
" 0.6265747 0.4375000 0.13125 0.6625000 \n",
" 0.6187487 0.4541667 0.13625 0.6663690 \n",
" 0.6175835 0.4458333 0.13375 0.6639881 \n",
" 0.5868027 0.4791667 0.14375 0.6663690 \n",
" 0.5746516 0.4916667 0.14750 0.6672619 \n",
" 0.6042179 0.4291667 0.12875 0.6547619 \n",
" 0.5858226 0.4708333 0.14125 0.6639881 \n",
" 0.5845809 0.4708333 0.14125 0.6630952 \n",
" 0.6265675 0.4250000 0.12750 0.6580357 \n",
" 0.6348284 0.4458333 0.13375 0.6666667 \n",
" 0.6280692 0.4541667 0.13625 0.6690476 \n",
" 0.5512584 0.4625000 0.13875 0.6482143 \n",
" 0.5748221 0.4416667 0.13250 0.6502976 \n",
" 0.5399437 0.4500000 0.13500 0.6428571 \n",
" 0.5898740 0.4750000 0.14250 0.6660714 \n",
" 0.5347326 0.4583333 0.13750 0.6434524 \n",
" 0.5558137 0.4958333 0.14875 0.6630952 \n",
" 0.6019034 0.4625000 0.13875 0.6651786 \n",
" 0.5934254 0.5041667 0.15125 0.6770833 \n",
" 0.5755710 0.4958333 0.14875 0.6675595 \n",
" 0.5338090 0.4583333 0.13750 0.6434524 \n",
" 0.5446951 0.4708333 0.14125 0.6505952 \n",
" 0.5453628 0.4833333 0.14500 0.6541667 \n",
" 0.6003662 0.4875000 0.14625 0.6723214 \n",
" 0.5972154 0.5083333 0.15250 0.6800595 \n",
" 0.5990483 0.5125000 0.15375 0.6821429 \n",
" 0.6096981 0.4833333 0.14500 0.6747024 \n",
" 0.5918111 0.5000000 0.15000 0.6758929 \n",
" 0.5685198 0.5083333 0.15250 0.6711310 \n",
" 0.5343391 0.5041667 0.15125 0.6565476 \n",
" 0.5019251 0.4625000 0.13875 0.6321429 \n",
" 0.5162603 0.4875000 0.14625 0.6455357 \n",
" 0.5755051 0.4916667 0.14750 0.6681548 \n",
" 0.5634594 0.4833333 0.14500 0.6604167 \n",
" 0.5381508 0.4750000 0.14250 0.6500000 \n",
" 0.5928798 0.4875000 0.14625 0.6678571 \n",
" 0.5959736 0.5000000 0.15000 0.6750000 \n",
" 0.5913461 0.5000000 0.15000 0.6732143 \n",
" 0.5193875 0.4875000 0.14625 0.6473214 \n",
" 0.5359917 0.4916667 0.14750 0.6547619 \n",
" 0.5522947 0.5166667 0.15500 0.6681548 \n",
" 0.5410904 0.4333333 0.13000 0.6372024 \n",
" 0.5297298 0.4625000 0.13875 0.6410714 \n",
" 0.5775526 0.4958333 0.14875 0.6657738 \n",
" 0.5555856 0.4500000 0.13500 0.6464286 \n",
" 0.5485550 0.4541667 0.13625 0.6458333 \n",
" 0.5541131 0.4500000 0.13500 0.6473214 \n",
"\n",
"Tuning parameter 'gamma' was held constant at a value of 0\n",
"Tuning\n",
" parameter 'min_child_weight' was held constant at a value of 1\n",
"Accuracy was used to select the optimal model using the largest value.\n",
"The final values used for the model were nrounds = 100, max_depth = 1, eta\n",
" = 0.3, gamma = 0, colsample_bytree = 0.8, min_child_weight = 1 and subsample\n",
" = 1.\n",
"\n",
"attr(,\"class\")\n",
"[1] \"caretList\""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"model_list1"
]
},
{
"cell_type": "code",
"execution_count": 25,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAA8AAAAJYCAYAAACtlI8BAAAEGWlDQ1BrQ0dDb2xvclNwYWNl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\nsh+aP9119AAAAABJRU5ErkJggg==",
"text/plain": [
"plot without title"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# Model Performance:\n",
"options(repr.plot.width = 8, repr.plot.height = 5)\n",
"total_df %>% \n",
" select(Accuracy, Precision, Recall, Sensitivity, Model) %>% \n",
" gather(a, b, -Model) %>% \n",
" ggplot(aes(Model, b, fill = Model, color = Model)) + \n",
" geom_boxplot(show.legend = FALSE, alpha = 0.3) + \n",
" facet_wrap(~ a, scales = \"free\") + \n",
" coord_flip() + \n",
" labs(x = NULL, y = NULL, \n",
" title = \"Evaluation and Comparison of Model Performance\")"
]
},
{
"cell_type": "code",
"execution_count": 26,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"\n",
"\n",
"|Model | Accuracy| Precision| Recall| Sensitivity|\n",
"|:--------|--------:|---------:|------:|-----------:|\n",
"|XgbTree | 75.50| 63.54| 43.75| 43.75|\n",
"|RF | 74.50| 61.27| 42.50| 42.50|\n",
"|Logistic | 74.12| 58.93| 46.25| 46.25|\n",
"|SVM | 70.88| 52.13| 29.58| 29.58|\n",
"|KNN | 66.88| 39.47| 16.67| 16.67|"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# table\n",
"options(warn=-1)\n",
"total_df %>% \n",
" group_by(Model) %>% \n",
" summarise_each(funs(mean), Accuracy, Precision, Recall, Sensitivity) %>% \n",
" ungroup() %>% \n",
" arrange(-Accuracy) %>% \n",
" mutate_if(is.numeric, function(x) {round(100*x, 2)}) %>% \n",
" knitr::kable(caption = \"Table 1: Model Performance for Validation Data Sets\")"
]
},
{
"cell_type": "code",
"execution_count": 27,
"metadata": {},
"outputs": [],
"source": [
"options(warn=-1)\n",
"total_df %>% \n",
" group_by(Model) %>% \n",
" summarise_each(funs(mean), Accuracy, Precision, Recall, Sensitivity) %>% \n",
" ungroup() %>% \n",
" arrange(-Accuracy) %>% \n",
" mutate_if(is.numeric, function(x) {round(100*x, 2)}) %>% \n",
" write.csv(., \"f1.csv\")"
]
},
{
"cell_type": "code",
"execution_count": 28,
"metadata": {},
"outputs": [],
"source": [
"#-----------------------------------------------------\n",
"# Confusion Matrix\n",
"#-----------------------------------------------------\n",
"pred <- predict(model_list1$xgbTree, test_data %>% select(-Class))\n",
"dfCM <- data.frame(pred = pred, act = test_data$Class, id = 1:200)"
]
},
{
"cell_type": "code",
"execution_count": 29,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
".\n",
" Bad Good \n",
" 0 156 "
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# \n",
"df1 <- dfCM %>% filter(pred == \"Good\")\n",
"\n",
"# number of loan contracts: \n",
"df1$pred %>% table()"
]
},
{
"cell_type": "code",
"execution_count": 30,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
".\n",
" Bad Good \n",
" 32 124 "
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# There are good cases = true positive and bad cases = false postive\n",
"df1$act %>% table()"
]
},
{
"cell_type": "code",
"execution_count": 31,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<table>\n",
"<thead><tr><th scope=col>pred</th><th scope=col>act</th><th scope=col>id</th></tr></thead>\n",
"<tbody>\n",
"\t<tr><td>Bad </td><td>Bad </td><td>1 </td></tr>\n",
"\t<tr><td>Bad </td><td>Bad </td><td>2 </td></tr>\n",
"\t<tr><td>Bad </td><td>Good</td><td>3 </td></tr>\n",
"\t<tr><td>Good</td><td>Good</td><td>4 </td></tr>\n",
"\t<tr><td>Good</td><td>Good</td><td>5 </td></tr>\n",
"</tbody>\n",
"</table>\n"
],
"text/latex": [
"\\begin{tabular}{r|lll}\n",
" pred & act & id\\\\\n",
"\\hline\n",
"\t Bad & Bad & 1 \\\\\n",
"\t Bad & Bad & 2 \\\\\n",
"\t Bad & Good & 3 \\\\\n",
"\t Good & Good & 4 \\\\\n",
"\t Good & Good & 5 \\\\\n",
"\\end{tabular}\n"
],
"text/markdown": [
"\n",
"| pred | act | id |\n",
"|---|---|---|\n",
"| Bad | Bad | 1 |\n",
"| Bad | Bad | 2 |\n",
"| Bad | Good | 3 |\n",
"| Good | Good | 4 |\n",
"| Good | Good | 5 |\n",
"\n"
],
"text/plain": [
" pred act id\n",
"1 Bad Bad 1 \n",
"2 Bad Bad 2 \n",
"3 Bad Good 3 \n",
"4 Good Good 4 \n",
"5 Good Good 5 "
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"head(dfCM,5)"
]
},
{
"cell_type": "code",
"execution_count": 32,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"Confusion Matrix and Statistics\n",
"\n",
" Reference\n",
"Prediction Bad Good\n",
" Bad 28 16\n",
" Good 32 124\n",
" \n",
" Accuracy : 0.76 \n",
" 95% CI : (0.6947, 0.8174)\n",
" No Information Rate : 0.7 \n",
" P-Value [Acc > NIR] : 0.03595 \n",
" \n",
" Kappa : 0.3814 \n",
" Mcnemar's Test P-Value : 0.03038 \n",
" \n",
" Sensitivity : 0.4667 \n",
" Specificity : 0.8857 \n",
" Pos Pred Value : 0.6364 \n",
" Neg Pred Value : 0.7949 \n",
" Prevalence : 0.3000 \n",
" Detection Rate : 0.1400 \n",
" Detection Prevalence : 0.2200 \n",
" Balanced Accuracy : 0.6762 \n",
" \n",
" 'Positive' Class : Bad \n",
" "
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# \n",
"confusionMatrix(dfCM$pred,dfCM$act)"
]
},
{
"cell_type": "code",
"execution_count": 33,
"metadata": {},
"outputs": [],
"source": [
"#----------------------------------------\n",
"# Profit Simulation\n",
"#----------------------------------------\n",
"\n",
"get_result <- function(ti_le, N, model_selected, df_test_selected) {\n",
" my_vec <- c()\n",
" \n",
" for (i in 1:N) {\n",
" set.seed(i) \n",
" df_test <- df_test_selected %>% \n",
" group_by(Class) %>% \n",
" sample_frac(ti_le) %>% \n",
" ungroup()\n",
" \n",
" pred <- predict(model_selected, df_test %>% select(-Class))\n",
" cm <- confusionMatrix(df_test$Class, pred)\n",
" cm$table %>% \n",
" as.vector() -> u\n",
" my_vec <- c(my_vec, u)\n",
" }\n",
" \n",
" my_vec %>% \n",
" matrix(ncol = 4, byrow = TRUE) %>% \n",
" as.data.frame() %>% \n",
" rename(BB = V1, \n",
" GB = V2,\n",
" BG = V3,\n",
" GG = V4) %>% \n",
" return()\n",
"}"
]
},
{
"cell_type": "code",
"execution_count": 34,
"metadata": {},
"outputs": [],
"source": [
"df_testa <- test_data %>% \n",
" group_by(Class) %>% \n",
" sample_frac(0.5) %>% ungroup()"
]
},
{
"cell_type": "code",
"execution_count": 35,
"metadata": {},
"outputs": [],
"source": [
"preda <- predict(model_list1$xgbTree, df_testa %>% select(-Class))"
]
},
{
"cell_type": "code",
"execution_count": 36,
"metadata": {},
"outputs": [],
"source": [
"cma <- confusionMatrix(df_testa$Class, preda)"
]
},
{
"cell_type": "code",
"execution_count": 37,
"metadata": {},
"outputs": [],
"source": [
"my_veca <- c()"
]
},
{
"cell_type": "code",
"execution_count": 38,
"metadata": {},
"outputs": [],
"source": [
"cma$table %>% \n",
" as.vector() -> u\n",
" my_veca <- c(my_veca, u)"
]
},
{
"cell_type": "code",
"execution_count": 39,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<table>\n",
"<thead><tr><th scope=col>BB</th><th scope=col>GB</th><th scope=col>BG</th><th scope=col>GG</th></tr></thead>\n",
"<tbody>\n",
"\t<tr><td>12</td><td>8 </td><td>18</td><td>62</td></tr>\n",
"</tbody>\n",
"</table>\n"
],
"text/latex": [
"\\begin{tabular}{r|llll}\n",
" BB & GB & BG & GG\\\\\n",
"\\hline\n",
"\t 12 & 8 & 18 & 62\\\\\n",
"\\end{tabular}\n"
],
"text/markdown": [
"\n",
"| BB | GB | BG | GG |\n",
"|---|---|---|---|\n",
"| 12 | 8 | 18 | 62 |\n",
"\n"
],
"text/plain": [
" BB GB BG GG\n",
"1 12 8 18 62"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
" my_veca %>% \n",
" matrix(ncol = 4, byrow = TRUE) %>% \n",
" as.data.frame() %>% \n",
" rename(BB = V1, \n",
" GB = V2,\n",
" BG = V3,\n",
" GG = V4)"
]
},
{
"cell_type": "code",
"execution_count": 40,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<ol class=list-inline>\n",
"\t<li>12</li>\n",
"\t<li>8</li>\n",
"\t<li>18</li>\n",
"\t<li>62</li>\n",
"</ol>\n"
],
"text/latex": [
"\\begin{enumerate*}\n",
"\\item 12\n",
"\\item 8\n",
"\\item 18\n",
"\\item 62\n",
"\\end{enumerate*}\n"
],
"text/markdown": [
"1. 12\n",
"2. 8\n",
"3. 18\n",
"4. 62\n",
"\n",
"\n"
],
"text/plain": [
"[1] 12 8 18 62"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"my_veca"
]
},
{
"cell_type": "code",
"execution_count": 41,
"metadata": {},
"outputs": [],
"source": [
"# result: \n",
"\n",
"loan_xgbtree <- get_result(0.5, 100, model_list1$xgbTree, test_data)\n",
"loan_svm <- get_result(0.5, 100, model_list1$svmLinear, test_data)\n",
"loan_rf <- get_result(0.5, 100, model_list1$rf, test_data)\n",
"loan_knn <- get_result(0.5, 100, model_list1$knn, test_data)\n",
"loan_logit <- get_result(0.5, 100, model_list1$glm, test_data)\n",
"\n",
"# BB = Bad Bad = True Negatives\n",
"# GB = Good Bad = False Negatives\n",
"# BG = Bad Good = False Positive\n",
"# GG = Good Good = True Positive\n"
]
},
{
"cell_type": "code",
"execution_count": 42,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"'data.frame':\t100 obs. of 4 variables:\n",
" $ BB: int 16 11 14 15 16 13 13 16 15 12 ...\n",
" $ GB: int 11 12 5 10 10 7 8 9 12 7 ...\n",
" $ BG: int 14 19 16 15 14 17 17 14 15 18 ...\n",
" $ GG: int 59 58 65 60 60 63 62 61 58 63 ...\n"
]
},
{
"data": {
"text/html": [
"<table>\n",
"<thead><tr><th scope=col>BB</th><th scope=col>GB</th><th scope=col>BG</th><th scope=col>GG</th></tr></thead>\n",
"<tbody>\n",
"\t<tr><td>16</td><td>11</td><td>14</td><td>59</td></tr>\n",
"\t<tr><td>11</td><td>12</td><td>19</td><td>58</td></tr>\n",
"\t<tr><td>14</td><td> 5</td><td>16</td><td>65</td></tr>\n",
"\t<tr><td>15</td><td>10</td><td>15</td><td>60</td></tr>\n",
"\t<tr><td>16</td><td>10</td><td>14</td><td>60</td></tr>\n",
"</tbody>\n",
"</table>\n"
],
"text/latex": [
"\\begin{tabular}{r|llll}\n",
" BB & GB & BG & GG\\\\\n",
"\\hline\n",
"\t 16 & 11 & 14 & 59\\\\\n",
"\t 11 & 12 & 19 & 58\\\\\n",
"\t 14 & 5 & 16 & 65\\\\\n",
"\t 15 & 10 & 15 & 60\\\\\n",
"\t 16 & 10 & 14 & 60\\\\\n",
"\\end{tabular}\n"
],
"text/markdown": [
"\n",
"| BB | GB | BG | GG |\n",
"|---|---|---|---|\n",
"| 16 | 11 | 14 | 59 |\n",
"| 11 | 12 | 19 | 58 |\n",
"| 14 | 5 | 16 | 65 |\n",
"| 15 | 10 | 15 | 60 |\n",
"| 16 | 10 | 14 | 60 |\n",
"\n"
],
"text/plain": [
" BB GB BG GG\n",
"1 16 11 14 59\n",
"2 11 12 19 58\n",
"3 14 5 16 65\n",
"4 15 10 15 60\n",
"5 16 10 14 60"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"str(loan_xgbtree)\n",
"head(loan_xgbtree,5)"
]
},
{
"cell_type": "code",
"execution_count": 43,
"metadata": {},
"outputs": [],
"source": [
"# Calculation of Accuracy: \n",
"total_df_result <- bind_rows(loan_xgbtree %>% mutate(Model = \"XgbTree\"), \n",
" loan_svm %>% mutate(Model = \"SVM\"), \n",
" loan_rf %>% mutate(Model = \"RF\"), \n",
" loan_knn %>% mutate(Model = \"KNN\"), \n",
" loan_logit %>% mutate(Model = \"Logistic\")) %>% \n",
" mutate(Accuracy = (BB + GG) / (BB + GG + GB + BG))"
]
},
{
"cell_type": "code",
"execution_count": 44,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<table>\n",
"<thead><tr><th scope=col>BB</th><th scope=col>GB</th><th scope=col>BG</th><th scope=col>GG</th><th scope=col>Model</th><th scope=col>Accuracy</th></tr></thead>\n",
"<tbody>\n",
"\t<tr><td>16 </td><td>11 </td><td>14 </td><td>59 </td><td>XgbTree</td><td>0.75 </td></tr>\n",
"\t<tr><td>11 </td><td>12 </td><td>19 </td><td>58 </td><td>XgbTree</td><td>0.69 </td></tr>\n",
"\t<tr><td>14 </td><td> 5 </td><td>16 </td><td>65 </td><td>XgbTree</td><td>0.79 </td></tr>\n",
"\t<tr><td>15 </td><td>10 </td><td>15 </td><td>60 </td><td>XgbTree</td><td>0.75 </td></tr>\n",
"\t<tr><td>16 </td><td>10 </td><td>14 </td><td>60 </td><td>XgbTree</td><td>0.76 </td></tr>\n",
"</tbody>\n",
"</table>\n"
],
"text/latex": [
"\\begin{tabular}{r|llllll}\n",
" BB & GB & BG & GG & Model & Accuracy\\\\\n",
"\\hline\n",
"\t 16 & 11 & 14 & 59 & XgbTree & 0.75 \\\\\n",
"\t 11 & 12 & 19 & 58 & XgbTree & 0.69 \\\\\n",
"\t 14 & 5 & 16 & 65 & XgbTree & 0.79 \\\\\n",
"\t 15 & 10 & 15 & 60 & XgbTree & 0.75 \\\\\n",
"\t 16 & 10 & 14 & 60 & XgbTree & 0.76 \\\\\n",
"\\end{tabular}\n"
],
"text/markdown": [
"\n",
"| BB | GB | BG | GG | Model | Accuracy |\n",
"|---|---|---|---|---|---|\n",
"| 16 | 11 | 14 | 59 | XgbTree | 0.75 |\n",
"| 11 | 12 | 19 | 58 | XgbTree | 0.69 |\n",
"| 14 | 5 | 16 | 65 | XgbTree | 0.79 |\n",
"| 15 | 10 | 15 | 60 | XgbTree | 0.75 |\n",
"| 16 | 10 | 14 | 60 | XgbTree | 0.76 |\n",
"\n"
],
"text/plain": [
" BB GB BG GG Model Accuracy\n",
"1 16 11 14 59 XgbTree 0.75 \n",
"2 11 12 19 58 XgbTree 0.69 \n",
"3 14 5 16 65 XgbTree 0.79 \n",
"4 15 10 15 60 XgbTree 0.75 \n",
"5 16 10 14 60 XgbTree 0.76 "
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"total_df_result %>% head(5)"
]
},
{
"cell_type": "code",
"execution_count": 45,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"\n",
"\n",
"|Model | Accuracy| GG| BB| BG|\n",
"|:--------|--------:|-----:|-----:|-----:|\n",
"|Logistic | 76.59| 60.36| 16.23| 13.77|\n",
"|XgbTree | 76.10| 61.83| 14.27| 15.73|\n",
"|RF | 75.54| 60.89| 14.65| 15.35|\n",
"|SVM | 74.35| 66.37| 7.98| 22.02|\n",
"|KNN | 69.23| 63.64| 5.59| 24.41|"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# Avg Accuracy from 100 samples: \n",
"options(warn=-1)\n",
"total_df_result %>% \n",
" group_by(Model) %>% \n",
" summarise_each(funs(mean), Accuracy, GG, BB, BG) %>% \n",
" ungroup() %>% \n",
" arrange(-Accuracy) %>% \n",
" mutate_at(.vars = c(\"Accuracy\"), function(x) {round(100*x, 2)}) %>% \n",
" knitr::kable(caption = \"Table 2: Model Performance for Test Data Sets\")"
]
},
{
"cell_type": "code",
"execution_count": 46,
"metadata": {},
"outputs": [],
"source": [
"# Profit simulation function: \n",
"profit_simu <- function(df_result, rate, N) {\n",
" data(\"GermanCredit\")\n",
" no_loan <- german$Amount\n",
" good_debt <- sum(df_result$GG)\n",
" bad_debt <- sum(df_result$BG)\n",
" \n",
" my_prof <- c()\n",
" \n",
" for (i in 1:N) {\n",
" set.seed(i)\n",
" prof <- rate*sample(no_loan, size = good_debt, replace = TRUE) %>% sum() \n",
" - sum(sample(no_loan, size = bad_debt, replace = TRUE))\n",
" my_prof <- c(prof, my_prof)\n",
" }\n",
" return(my_prof)\n",
"}"
]
},
{
"cell_type": "code",
"execution_count": 47,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"1000"
],
"text/latex": [
"1000"
],
"text/markdown": [
"1000"
],
"text/plain": [
"[1] 1000"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"6183"
],
"text/latex": [
"6183"
],
"text/markdown": [
"6183"
],
"text/plain": [
"[1] 6183"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"data(\"GermanCredit\")\n",
"length(GermanCredit$Amount)\n",
"sum(loan_xgbtree$GG)"
]
},
{
"cell_type": "code",
"execution_count": 48,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<ol class=list-inline>\n",
"\t<li>59</li>\n",
"\t<li>58</li>\n",
"\t<li>65</li>\n",
"\t<li>60</li>\n",
"\t<li>60</li>\n",
"\t<li>63</li>\n",
"\t<li>62</li>\n",
"\t<li>61</li>\n",
"\t<li>58</li>\n",
"\t<li>63</li>\n",
"\t<li>64</li>\n",
"\t<li>62</li>\n",
"\t<li>63</li>\n",
"\t<li>62</li>\n",
"\t<li>61</li>\n",
"\t<li>67</li>\n",
"\t<li>62</li>\n",
"\t<li>60</li>\n",
"\t<li>62</li>\n",
"\t<li>61</li>\n",
"\t<li>64</li>\n",
"\t<li>63</li>\n",
"\t<li>61</li>\n",
"\t<li>59</li>\n",
"\t<li>64</li>\n",
"\t<li>63</li>\n",
"\t<li>65</li>\n",
"\t<li>64</li>\n",
"\t<li>62</li>\n",
"\t<li>61</li>\n",
"\t<li>61</li>\n",
"\t<li>58</li>\n",
"\t<li>63</li>\n",
"\t<li>61</li>\n",
"\t<li>62</li>\n",
"\t<li>60</li>\n",
"\t<li>59</li>\n",
"\t<li>63</li>\n",
"\t<li>61</li>\n",
"\t<li>62</li>\n",
"\t<li>63</li>\n",
"\t<li>58</li>\n",
"\t<li>61</li>\n",
"\t<li>62</li>\n",
"\t<li>61</li>\n",
"\t<li>62</li>\n",
"\t<li>64</li>\n",
"\t<li>64</li>\n",
"\t<li>60</li>\n",
"\t<li>58</li>\n",
"\t<li>62</li>\n",
"\t<li>63</li>\n",
"\t<li>63</li>\n",
"\t<li>61</li>\n",
"\t<li>63</li>\n",
"\t<li>62</li>\n",
"\t<li>58</li>\n",
"\t<li>62</li>\n",
"\t<li>63</li>\n",
"\t<li>58</li>\n",
"\t<li>64</li>\n",
"\t<li>65</li>\n",
"\t<li>64</li>\n",
"\t<li>65</li>\n",
"\t<li>61</li>\n",
"\t<li>65</li>\n",
"\t<li>61</li>\n",
"\t<li>62</li>\n",
"\t<li>64</li>\n",
"\t<li>63</li>\n",
"\t<li>60</li>\n",
"\t<li>60</li>\n",
"\t<li>65</li>\n",
"\t<li>65</li>\n",
"\t<li>61</li>\n",
"\t<li>61</li>\n",
"\t<li>60</li>\n",
"\t<li>59</li>\n",
"\t<li>64</li>\n",
"\t<li>61</li>\n",
"\t<li>59</li>\n",
"\t<li>63</li>\n",
"\t<li>60</li>\n",
"\t<li>60</li>\n",
"\t<li>61</li>\n",
"\t<li>65</li>\n",
"\t<li>62</li>\n",
"\t<li>62</li>\n",
"\t<li>63</li>\n",
"\t<li>60</li>\n",
"\t<li>64</li>\n",
"\t<li>62</li>\n",
"\t<li>61</li>\n",
"\t<li>61</li>\n",
"\t<li>61</li>\n",
"\t<li>64</li>\n",
"\t<li>61</li>\n",
"\t<li>60</li>\n",
"\t<li>64</li>\n",
"\t<li>63</li>\n",
"</ol>\n"
],
"text/latex": [
"\\begin{enumerate*}\n",
"\\item 59\n",
"\\item 58\n",
"\\item 65\n",
"\\item 60\n",
"\\item 60\n",
"\\item 63\n",
"\\item 62\n",
"\\item 61\n",
"\\item 58\n",
"\\item 63\n",
"\\item 64\n",
"\\item 62\n",
"\\item 63\n",
"\\item 62\n",
"\\item 61\n",
"\\item 67\n",
"\\item 62\n",
"\\item 60\n",
"\\item 62\n",
"\\item 61\n",
"\\item 64\n",
"\\item 63\n",
"\\item 61\n",
"\\item 59\n",
"\\item 64\n",
"\\item 63\n",
"\\item 65\n",
"\\item 64\n",
"\\item 62\n",
"\\item 61\n",
"\\item 61\n",
"\\item 58\n",
"\\item 63\n",
"\\item 61\n",
"\\item 62\n",
"\\item 60\n",
"\\item 59\n",
"\\item 63\n",
"\\item 61\n",
"\\item 62\n",
"\\item 63\n",
"\\item 58\n",
"\\item 61\n",
"\\item 62\n",
"\\item 61\n",
"\\item 62\n",
"\\item 64\n",
"\\item 64\n",
"\\item 60\n",
"\\item 58\n",
"\\item 62\n",
"\\item 63\n",
"\\item 63\n",
"\\item 61\n",
"\\item 63\n",
"\\item 62\n",
"\\item 58\n",
"\\item 62\n",
"\\item 63\n",
"\\item 58\n",
"\\item 64\n",
"\\item 65\n",
"\\item 64\n",
"\\item 65\n",
"\\item 61\n",
"\\item 65\n",
"\\item 61\n",
"\\item 62\n",
"\\item 64\n",
"\\item 63\n",
"\\item 60\n",
"\\item 60\n",
"\\item 65\n",
"\\item 65\n",
"\\item 61\n",
"\\item 61\n",
"\\item 60\n",
"\\item 59\n",
"\\item 64\n",
"\\item 61\n",
"\\item 59\n",
"\\item 63\n",
"\\item 60\n",
"\\item 60\n",
"\\item 61\n",
"\\item 65\n",
"\\item 62\n",
"\\item 62\n",
"\\item 63\n",
"\\item 60\n",
"\\item 64\n",
"\\item 62\n",
"\\item 61\n",
"\\item 61\n",
"\\item 61\n",
"\\item 64\n",
"\\item 61\n",
"\\item 60\n",
"\\item 64\n",
"\\item 63\n",
"\\end{enumerate*}\n"
],
"text/markdown": [
"1. 59\n",
"2. 58\n",
"3. 65\n",
"4. 60\n",
"5. 60\n",
"6. 63\n",
"7. 62\n",
"8. 61\n",
"9. 58\n",
"10. 63\n",
"11. 64\n",
"12. 62\n",
"13. 63\n",
"14. 62\n",
"15. 61\n",
"16. 67\n",
"17. 62\n",
"18. 60\n",
"19. 62\n",
"20. 61\n",
"21. 64\n",
"22. 63\n",
"23. 61\n",
"24. 59\n",
"25. 64\n",
"26. 63\n",
"27. 65\n",
"28. 64\n",
"29. 62\n",
"30. 61\n",
"31. 61\n",
"32. 58\n",
"33. 63\n",
"34. 61\n",
"35. 62\n",
"36. 60\n",
"37. 59\n",
"38. 63\n",
"39. 61\n",
"40. 62\n",
"41. 63\n",
"42. 58\n",
"43. 61\n",
"44. 62\n",
"45. 61\n",
"46. 62\n",
"47. 64\n",
"48. 64\n",
"49. 60\n",
"50. 58\n",
"51. 62\n",
"52. 63\n",
"53. 63\n",
"54. 61\n",
"55. 63\n",
"56. 62\n",
"57. 58\n",
"58. 62\n",
"59. 63\n",
"60. 58\n",
"61. 64\n",
"62. 65\n",
"63. 64\n",
"64. 65\n",
"65. 61\n",
"66. 65\n",
"67. 61\n",
"68. 62\n",
"69. 64\n",
"70. 63\n",
"71. 60\n",
"72. 60\n",
"73. 65\n",
"74. 65\n",
"75. 61\n",
"76. 61\n",
"77. 60\n",
"78. 59\n",
"79. 64\n",
"80. 61\n",
"81. 59\n",
"82. 63\n",
"83. 60\n",
"84. 60\n",
"85. 61\n",
"86. 65\n",
"87. 62\n",
"88. 62\n",
"89. 63\n",
"90. 60\n",
"91. 64\n",
"92. 62\n",
"93. 61\n",
"94. 61\n",
"95. 61\n",
"96. 64\n",
"97. 61\n",
"98. 60\n",
"99. 64\n",
"100. 63\n",
"\n",
"\n"
],
"text/plain": [
" [1] 59 58 65 60 60 63 62 61 58 63 64 62 63 62 61 67 62 60 62 61 64 63 61 59 64\n",
" [26] 63 65 64 62 61 61 58 63 61 62 60 59 63 61 62 63 58 61 62 61 62 64 64 60 58\n",
" [51] 62 63 63 61 63 62 58 62 63 58 64 65 64 65 61 65 61 62 64 63 60 60 65 65 61\n",
" [76] 61 60 59 64 61 59 63 60 60 61 65 62 62 63 60 64 62 61 61 61 64 61 60 64 63"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"loan_xgbtree$GG"
]
},
{
"cell_type": "code",
"execution_count": 49,
"metadata": {},
"outputs": [],
"source": [
"# Profit Calculation from 1000 simulation times\n",
"\n",
"profit <- c(loan_xgbtree %>% profit_simu(0.3, 1000), \n",
" loan_svm %>% profit_simu(0.3, 1000), \n",
" loan_rf %>% profit_simu(0.3, 1000), \n",
" loan_knn %>% profit_simu(0.3, 1000), \n",
" loan_logit %>% profit_simu(0.3, 1000))\n",
"\n",
"profit_df <- data.frame(Profit = profit, \n",
" Model = c(rep(\"XgbTree\", 1000), \n",
" rep(\"SVM\", 1000), \n",
" rep(\"RF\", 1000), \n",
" rep(\"KNN\", 1000), \n",
" rep(\"Logistic\", 1000)\n",
" )\n",
" )"
]
},
{
"cell_type": "code",
"execution_count": 50,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<table>\n",
"<thead><tr><th scope=col>Profit</th><th scope=col>Model</th></tr></thead>\n",
"<tbody>\n",
"\t<tr><td>6072775</td><td>XgbTree</td></tr>\n",
"\t<tr><td>6086358</td><td>XgbTree</td></tr>\n",
"\t<tr><td>6037586</td><td>XgbTree</td></tr>\n",
"\t<tr><td>6102285</td><td>XgbTree</td></tr>\n",
"\t<tr><td>6065468</td><td>XgbTree</td></tr>\n",
"</tbody>\n",
"</table>\n"
],
"text/latex": [
"\\begin{tabular}{r|ll}\n",
" Profit & Model\\\\\n",
"\\hline\n",
"\t 6072775 & XgbTree\\\\\n",
"\t 6086358 & XgbTree\\\\\n",
"\t 6037586 & XgbTree\\\\\n",
"\t 6102285 & XgbTree\\\\\n",
"\t 6065468 & XgbTree\\\\\n",
"\\end{tabular}\n"
],
"text/markdown": [
"\n",
"| Profit | Model |\n",
"|---|---|\n",
"| 6072775 | XgbTree |\n",
"| 6086358 | XgbTree |\n",
"| 6037586 | XgbTree |\n",
"| 6102285 | XgbTree |\n",
"| 6065468 | XgbTree |\n",
"\n"
],
"text/plain": [
" Profit Model \n",
"1 6072775 XgbTree\n",
"2 6086358 XgbTree\n",
"3 6037586 XgbTree\n",
"4 6102285 XgbTree\n",
"5 6065468 XgbTree"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"head(profit_df,5)"
]
},
{
"cell_type": "code",
"execution_count": 51,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"\n",
"\n",
"|Model | Mean| Median| Min| Max| SD|\n",
"|:--------|-------:|-------:|-------:|-------:|-----:|\n",
"|SVM | 6513828| 6514225| 6286726| 6716445| 68135|\n",
"|KNN | 6246711| 6246003| 6027972| 6440565| 66505|\n",
"