Praktikum 6
Nurul Fadila U
2025-04-21
Library
library(readxl)
library(e1071) ## Warning: package 'e1071' was built under R version 4.4.3library(caret) ## Warning: package 'caret' was built under R version 4.4.3## Loading required package: ggplot2## Loading required package: latticelibrary(class)
library(Metrics)## Warning: package 'Metrics' was built under R version 4.4.3##
## Attaching package: 'Metrics'## The following objects are masked from 'package:caret':
##
## precision, recalllibrary(MLmetrics)##
## Attaching package: 'MLmetrics'## The following objects are masked from 'package:caret':
##
## MAE, RMSE## The following object is masked from 'package:base':
##
## Recalllibrary(party)## Loading required package: grid## Loading required package: mvtnorm## Loading required package: modeltools## Loading required package: stats4## Loading required package: strucchange## Loading required package: zoo##
## Attaching package: 'zoo'## The following objects are masked from 'package:base':
##
## as.Date, as.Date.numeric## Loading required package: sandwichlibrary(neuralnet)## Warning: package 'neuralnet' was built under R version 4.4.3library(randomForest)## Warning: package 'randomForest' was built under R version 4.4.3## randomForest 4.7-1.2## Type rfNews() to see new features/changes/bug fixes.##
## Attaching package: 'randomForest'## The following object is masked from 'package:ggplot2':
##
## marginlibrary(rpart)## Warning: package 'rpart' was built under R version 4.4.3library(rpart.plot)## Warning: package 'rpart.plot' was built under R version 4.4.3Input data
#input data
data <- read_excel("hasil_prepo_bank.xlsx")
print(data)## # A tibble: 4,521 × 17
## Age job marital education default balance housing loan contact day
## <dbl> <chr> <chr> <chr> <chr> <dbl> <chr> <chr> <chr> <dbl>
## 1 30 unemploy… married primary no 1787 no no cellul… 19
## 2 33 services married secondary no 4789 yes yes cellul… 11
## 3 35 manageme… single tertiary no 1350 yes no cellul… 16
## 4 30 manageme… married tertiary no 1476 yes yes unknown 3
## 5 59 blue-col… married secondary no 0 yes no unknown 5
## 6 35 manageme… single tertiary no 747 no no cellul… 23
## 7 36 self-emp… married tertiary no 307 yes no cellul… 14
## 8 39 technici… married secondary no 147 yes no cellul… 6
## 9 41 entrepre… married tertiary no 221 yes no unknown 14
## 10 43 services married primary no -88 yes yes cellul… 17
## # ℹ 4,511 more rows
## # ℹ 7 more variables: month <chr>, duration <dbl>, campaign <dbl>, pdays <dbl>,
## # previous <dbl>, poutcome <chr>, y <chr>Partisi data
str(data)## tibble [4,521 × 17] (S3: tbl_df/tbl/data.frame)
## $ Age : num [1:4521] 30 33 35 30 59 35 36 39 41 43 ...
## $ job : chr [1:4521] "unemployed" "services" "management" "management" ...
## $ marital : chr [1:4521] "married" "married" "single" "married" ...
## $ education: chr [1:4521] "primary" "secondary" "tertiary" "tertiary" ...
## $ default : chr [1:4521] "no" "no" "no" "no" ...
## $ balance : num [1:4521] 1787 4789 1350 1476 0 ...
## $ housing : chr [1:4521] "no" "yes" "yes" "yes" ...
## $ loan : chr [1:4521] "no" "yes" "no" "yes" ...
## $ contact : chr [1:4521] "cellular" "cellular" "cellular" "unknown" ...
## $ day : num [1:4521] 19 11 16 3 5 23 14 6 14 17 ...
## $ month : chr [1:4521] "oct" "may" "apr" "jun" ...
## $ duration : num [1:4521] 79 220 185 199 226 141 341 151 57 313 ...
## $ campaign : num [1:4521] 1 1 1 4 1 2 1 2 2 1 ...
## $ pdays : num [1:4521] -1 339 330 -1 -1 176 330 -1 -1 147 ...
## $ previous : num [1:4521] 0 4 1 0 0 3 2 0 0 2 ...
## $ poutcome : chr [1:4521] "unknown" "failure" "failure" "unknown" ...
## $ y : chr [1:4521] "no" "no" "no" "no" ...Sample <- sample(1:4521, 904)
testing <- data[Sample, ]
learning <- data[-Sample, ]NB
#Membuat model
model <- naiveBayes(y ~ ., data = learning)Prediksi dan evaluasi data training
prediksi <- predict(model, learning)
predicted <- prediksi
actual <- learning$y
actual <- ifelse(as.character(actual) == "yes", 0, 1)
predicted <- ifelse(as.character(predicted) == "yes", 0, 1)
# Generate confusion matrix
xtab <- table(actual, predicted)
cm <- caret::confusionMatrix(xtab)
print(cm)## Confusion Matrix and Statistics
##
## predicted
## actual 0 1
## 0 199 199
## 1 282 2937
##
## Accuracy : 0.867
## 95% CI : (0.8555, 0.8779)
## No Information Rate : 0.867
## P-Value [Acc > NIR] : 0.5121545
##
## Kappa : 0.3779
##
## Mcnemar's Test P-Value : 0.0001848
##
## Sensitivity : 0.41372
## Specificity : 0.93654
## Pos Pred Value : 0.50000
## Neg Pred Value : 0.91240
## Prevalence : 0.13298
## Detection Rate : 0.05502
## Detection Prevalence : 0.11004
## Balanced Accuracy : 0.67513
##
## 'Positive' Class : 0
## #Menghitung akurasi
#akurasi <- mean(prediksi == learning$y)
#print(paste("Akurasi_train:", round(akurasi * 100, 2), "%"))
accuracy <- (sum(diag(cm$table)))/sum(cm$table)
print(paste("Akurasi_train:", round(accuracy * 100, 2), "%"))## [1] "Akurasi_train: 86.7 %"# Compute Recall (correct order: actual, predicted)
recall_score <- recall(actual, predicted)
# Compute Precision (correct order: actual, predicted)
precision_score <- precision(actual, predicted)
# Compute F1-Score (correct order: actual, predicted)
#f1_score <- f1(actual, predicted)
f1_score <- 2 * ((precision_score * recall_score) / (precision_score + recall_score))
# Print results
cat("Recall:", recall_score, "\n")## Recall: 0.9123952cat("Precision:", precision_score, "\n")## Precision: 0.9365434cat("F1-Score:", f1_score, "\n")## F1-Score: 0.9243116Prediksi dan evaluasi data testing
prediksi <- predict(model, testing)
predicted <- prediksi
actual <- testing$y
actual <- ifelse(as.character(actual) == "yes", 0, 1)
predicted <- ifelse(as.character(predicted) == "yes", 0, 1)
# Generate confusion matrix
xtab <- table(actual, predicted)
cm <- caret::confusionMatrix(xtab)
print(cm)## Confusion Matrix and Statistics
##
## predicted
## actual 0 1
## 0 68 55
## 1 78 703
##
## Accuracy : 0.8529
## 95% CI : (0.8281, 0.8753)
## No Information Rate : 0.8385
## P-Value [Acc > NIR] : 0.12864
##
## Kappa : 0.4199
##
## Mcnemar's Test P-Value : 0.05644
##
## Sensitivity : 0.46575
## Specificity : 0.92744
## Pos Pred Value : 0.55285
## Neg Pred Value : 0.90013
## Prevalence : 0.16150
## Detection Rate : 0.07522
## Detection Prevalence : 0.13606
## Balanced Accuracy : 0.69660
##
## 'Positive' Class : 0
## #Menghitung akurasi
#akurasi <- mean(prediksi == learning$y)
#print(paste("Akurasi_train:", round(akurasi * 100, 2), "%"))
accuracy <- (sum(diag(cm$table)))/sum(cm$table)
print(paste("Akurasi_test:", round(accuracy * 100, 2), "%"))## [1] "Akurasi_test: 85.29 %"# Compute Recall (correct order: actual, predicted)
recall_score <- recall(actual, predicted)
# Compute Precision (correct order: actual, predicted)
precision_score <- precision(actual, predicted)
# Compute F1-Score (correct order: actual, predicted)
#f1_score <- f1(actual, predicted)
f1_score <- 2 * ((precision_score * recall_score) / (precision_score + recall_score))
# Print results
cat("Recall:", recall_score, "\n")## Recall: 0.900128cat("Precision:", precision_score, "\n")## Precision: 0.9274406cat("F1-Score:", f1_score, "\n")## F1-Score: 0.9135802Decision Tree
data <- read_excel("hasil_prepo_bank.xlsx")
# Ubah semua kolom character menjadi factor
data[] <- lapply(data, function(x) if (is.character(x)) as.factor(x) else x)
Sample <- sample(1:4521, 904)
testing <- data[Sample, ]
learning <- data[-Sample, ]
# Bangun model pohon keputusan dengan ctree
output.tree <- ctree(y ~ ., data = learning)
# Plot hasil pohon keputusan
plot(output.tree)
# Prediksi dan evaluasi data training
prediksi <- predict(output.tree, learning)
predicted <- prediksi
actual <- learning$y
actual <- ifelse(as.character(actual) == "yes", 0, 1)
predicted <- ifelse(as.character(predicted) == "yes", 0, 1)
CM <- table(learning$y, prediksi)
CM## prediksi
## no yes
## no 2983 209
## yes 156 269accuracy <- (sum(diag(CM)))/sum(CM)
accuracy## [1] 0.8990876# Compute Recall (correct order: actual, predicted)
recall_score <- recall(actual, predicted)
# Compute Precision (correct order: actual, predicted)
precision_score <- precision(actual, predicted)
# Compute F1-Score (correct order: actual, predicted)
#f1_score <- f1(actual, predicted)
f1_score <- 2 * ((precision_score * recall_score) / (precision_score + recall_score))
# Print results
cat("Recall:", recall_score, "\n")## Recall: 0.9345238cat("Precision:", precision_score, "\n")## Precision: 0.9503026cat("F1-Score:", f1_score, "\n")## F1-Score: 0.9423472Prediksi dan evaluasi data testing
prediksi <- predict(output.tree, testing)
predicted <- prediksi
actual <- testing$y
actual <- ifelse(as.character(actual) == "yes", 0, 1)
predicted <- ifelse(as.character(predicted) == "yes", 0, 1)
CM <- table(testing$y, prediksi)
CM## prediksi
## no yes
## no 746 62
## yes 34 62accuracy <- (sum(diag(CM)))/sum(CM)
accuracy## [1] 0.8938053# Compute Recall (correct order: actual, predicted)
recall_score <- recall(actual, predicted)
# Compute Precision (correct order: actual, predicted)
precision_score <- precision(actual, predicted)
# Compute F1-Score (correct order: actual, predicted)
#f1_score <- f1(actual, predicted)
f1_score <- 2 * ((precision_score * recall_score) / (precision_score + recall_score))
# Print results
cat("Recall:", recall_score, "\n")## Recall: 0.9232673cat("Precision:", precision_score, "\n")## Precision: 0.9564103cat("F1-Score:", f1_score, "\n")## F1-Score: 0.9395466Visualisasi pohon keputusan
library(rpart.plot)
model <- rpart(y ~ ., data = learning, method = "class")
rpart.plot(model, main = "Pohon Keputusan")
KNN
data <- read_excel("hasil_prepo_bank.xlsx")
data$job <- as.factor(data$job)
data$marital <- as.factor(data$marital)
data$education <- as.factor(data$education)
data$default <- as.factor(data$default)
data$housing <- as.factor(data$housing)
data$loan <- as.factor(data$loan)
data$contact <- as.factor(data$contact)
data$month <- as.factor(data$month)
data$poutcome <- as.factor(data$poutcome)
data$y <- as.factor(data$y)
str(data)## tibble [4,521 × 17] (S3: tbl_df/tbl/data.frame)
## $ Age : num [1:4521] 30 33 35 30 59 35 36 39 41 43 ...
## $ job : Factor w/ 12 levels "admin.","blue-collar",..: 11 8 5 5 2 5 7 10 3 8 ...
## $ marital : Factor w/ 3 levels "divorced","married",..: 2 2 3 2 2 3 2 2 2 2 ...
## $ education: Factor w/ 4 levels "primary","secondary",..: 1 2 3 3 2 3 3 2 3 1 ...
## $ default : Factor w/ 2 levels "no","yes": 1 1 1 1 1 1 1 1 1 1 ...
## $ balance : num [1:4521] 1787 4789 1350 1476 0 ...
## $ housing : Factor w/ 2 levels "no","yes": 1 2 2 2 2 1 2 2 2 2 ...
## $ loan : Factor w/ 2 levels "no","yes": 1 2 1 2 1 1 1 1 1 2 ...
## $ contact : Factor w/ 3 levels "cellular","telephone",..: 1 1 1 3 3 1 1 1 3 1 ...
## $ day : num [1:4521] 19 11 16 3 5 23 14 6 14 17 ...
## $ month : Factor w/ 12 levels "apr","aug","dec",..: 11 9 1 7 9 4 9 9 9 1 ...
## $ duration : num [1:4521] 79 220 185 199 226 141 341 151 57 313 ...
## $ campaign : num [1:4521] 1 1 1 4 1 2 1 2 2 1 ...
## $ pdays : num [1:4521] -1 339 330 -1 -1 176 330 -1 -1 147 ...
## $ previous : num [1:4521] 0 4 1 0 0 3 2 0 0 2 ...
## $ poutcome : Factor w/ 4 levels "failure","other",..: 4 1 1 4 4 1 2 4 4 1 ...
## $ y : Factor w/ 2 levels "no","yes": 1 1 1 1 1 1 1 1 1 1 ...fitur_numeric <- data[, sapply(data, is.numeric)]
normalize <- function(x) { (x - min(x)) / (max(x) - min(x)) }
fitur_norm <- as.data.frame(lapply(fitur_numeric, normalize))
fitur_norm <- na.omit(fitur_norm)
target <- as.factor(data$y)
fitur_norm$y <- target
fitur_norm <- na.omit(fitur_norm)
target <- fitur_norm$y
fitur_norm$y <- NULL
fitur_final <- fitur_norm
set.seed(123)
inTrain <- createDataPartition(target, p = 0.8, list = FALSE)
trainData <- fitur_final[inTrain, ]
testData <- fitur_final[-inTrain, ]
trainLabel <- target[inTrain]
testLabel <- target[-inTrain]
trainData <- as.data.frame(lapply(trainData, as.numeric))
testData <- as.data.frame(lapply(testData, as.numeric))
train_dummy <- trainData
test_dummy <- testDataMenetukan nilai K optimal
library(class)
# Buat vector kosong buat simpan akurasi
k_values <- 1:10
accuracy_k <- numeric(length(k_values))
# Loop buat cari akurasi tiap k
for (i in k_values) {
pred_k <- knn(train = train_dummy, test = test_dummy, cl = trainLabel, k = i)
acc <- confusionMatrix(pred_k, testLabel, positive = "yes")$overall['Accuracy']
accuracy_k[i] <- acc
}
# Cari k dengan akurasi terbaik
best_k <- which.max(accuracy_k)
k <- best_k
cat("k terbaik adalah:", best_k, "dengan akurasi", round(accuracy_k[best_k] * 100, 2), "%\n")## k terbaik adalah: 8 dengan akurasi 90.04 %# Visualisasi akurasi k terbaik
plot(k_values, accuracy_k, type = "b", pch = 19, col = "blue",
xlab = "Nilai k", ylab = "Akurasi", main = "Cari k Terbaik buat kNN")
abline(v = best_k, col = "red", lty = 2)
Pembuatan model KNN serta prediksi data training dan testing
pred_train <- knn(trainData, trainData, cl = trainLabel, k = k)
cm_train <- confusionMatrix(pred_train, trainLabel, positive = "no")
pred_test <- knn(trainData, testData, cl = trainLabel, k = k)
cm_test <- confusionMatrix(pred_test, testLabel, positive = "no")Metrik Evaluasi
accuracy_knn_train <- cm_train$overall["Accuracy"]
precision_knn_train <- cm_train$byClass["Precision"]
recall_knn_train <- cm_train$byClass["Recall"]
f1_knn_train <- cm_train$byClass["F1"]
accuracy_knn_test <- cm_test$overall["Accuracy"]
precision_knn_test <- cm_test$byClass["Precision"]
recall_knn_test <- cm_test$byClass["Recall"]
f1_knn_test <- cm_test$byClass["F1"]
cat("\nConfusion Matrix data training (80:20)\n")##
## Confusion Matrix data training (80:20)print(cm_train$table)## Reference
## Prediction no yes
## no 3151 314
## yes 49 103cat("\nConfusion Matrix data testing (80:20)\n")##
## Confusion Matrix data testing (80:20)print(cm_test$table)## Reference
## Prediction no yes
## no 787 81
## yes 13 23cat("=== Hasil Evaluasi k-Nearest Neighbor ===\n\n")## === Hasil Evaluasi k-Nearest Neighbor ===cat("Data Training:\n")## Data Training:cat("Akurasi :", round(accuracy_knn_train * 100, 2), "%\n")## Akurasi : 89.96 %cat("Presisi :", round(precision_knn_train * 100, 2), "%\n")## Presisi : 90.94 %cat("Recall :", round(recall_knn_train * 100, 2), "%\n")## Recall : 98.47 %cat("F1 Score :", round(f1_knn_train * 100, 2), "%\n\n")## F1 Score : 94.55 %cat("Data Testing:\n")## Data Testing:cat("Akurasi :", round(accuracy_knn_test * 100, 2), "%\n")## Akurasi : 89.6 %cat("Presisi :", round(precision_knn_test * 100, 2), "%\n")## Presisi : 90.67 %cat("Recall :", round(recall_knn_test * 100, 2), "%\n")## Recall : 98.38 %cat("F1 Score :", round(f1_knn_test * 100, 2), "%\n")## F1 Score : 94.36 %Random Forest
data$job <- as.factor(data$job)
data$marital <- as.factor(data$marital)
data$education <- as.factor(data$education)
data$default <- as.factor(data$default)
data$housing <- as.factor(data$housing)
data$loan <- as.factor(data$loan)
data$contact <- as.factor(data$contact)data$month <- as.factor(data$month)
data$poutcome <- as.factor(data$poutcome)
data$y <- as.factor(data$y)
set.seed(123)
# Bagi data 80% training, 20% testing
inTrain <- createDataPartition(data$y, p = 0.8, list = FALSE)
trainData <- data[inTrain, ]
testData <- data[-inTrain, ]Pembentukan Model Random Forest
model_rf <- randomForest(y ~ ., data = trainData,
ntree = 100,
mtry = 2,
maxnodes = 30,
importance = TRUE)Metrik evaluasi
# Prediksi data training dan testing
pred_train_rf <- predict(model_rf, trainData)
pred_test_rf <- predict(model_rf, testData)
# Confusion Matrix
conf_train_rf <- confusionMatrix(pred_train_rf, trainData$y, positive = "no")
conf_test_rf <- confusionMatrix(pred_test_rf, testData$y, positive = "no")
# Metrik Evaluasi
accuracy_rf_train <- conf_train_rf$overall['Accuracy']
precision_rf_train <- conf_train_rf$byClass['Precision']
recall_rf_train <- conf_train_rf$byClass['Recall']
f1_rf_train <- F1_Score(y_pred = pred_train_rf, y_true = trainData$y, positive = "no")
accuracy_rf_test <- conf_test_rf$overall['Accuracy']
precision_rf_test <- conf_test_rf$byClass['Precision']
recall_rf_test <- conf_test_rf$byClass['Recall']
f1_rf_test <- F1_Score(y_pred = pred_test_rf, y_true = testData$y, positive = "no")
cat("\nConfusion Matrix data training (80:20)\n")##
## Confusion Matrix data training (80:20)print(conf_train_rf$table)## Reference
## Prediction no yes
## no 3200 371
## yes 0 46cat("\nConfusion Matrix data testing (80:20)\n")##
## Confusion Matrix data testing (80:20)print(conf_test_rf$table)## Reference
## Prediction no yes
## no 798 100
## yes 2 4cat("=== Hasil Evaluasi Random Forest ===\n\n")## === Hasil Evaluasi Random Forest ===cat("Data Training:\n")## Data Training:cat("Akurasi :", round(accuracy_rf_train * 100, 2), "%\n")## Akurasi : 89.74 %cat("Presisi :", round(precision_rf_train * 100, 2), "%\n")## Presisi : 89.61 %cat("Recall :", round(recall_rf_train * 100, 2), "%\n")## Recall : 100 %cat("F1 Score :", round(f1_rf_train * 100, 2), "%\n\n")## F1 Score : 94.52 %cat("Data Testing:\n")## Data Testing:cat("Akurasi :", round(accuracy_rf_test * 100, 2), "%\n")## Akurasi : 88.72 %cat("Presisi :", round(precision_rf_test * 100, 2), "%\n")## Presisi : 88.86 %cat("Recall :", round(recall_rf_test * 100, 2), "%\n")## Recall : 99.75 %cat("F1 Score :", round(f1_rf_test * 100, 2), "%\n")## F1 Score : 93.99 %ANN
data <- read_excel("hasil_resampling_bank.xlsx")
data$job <- as.factor(data$job)
data$marital <- as.factor(data$marital)
data$education <- as.factor(data$education)
data$default <- as.factor(data$default)
data$housing <- as.factor(data$housing)
data$loan <- as.factor(data$loan)
data$contact <- as.factor(data$contact)
data$month <- as.factor(data$month)
data$poutcome <- as.factor(data$poutcome)
data$y <- as.factor(data$y)
head(data)str(data)## tibble [8,000 × 17] (S3: tbl_df/tbl/data.frame)
## $ Age : num [1:8000] 30 33 35 30 59 35 36 39 41 43 ...
## $ job : Factor w/ 12 levels "admin.","blue-collar",..: 11 8 5 5 2 5 7 10 3 8 ...
## $ marital : Factor w/ 3 levels "divorced","married",..: 2 2 3 2 2 3 2 2 2 2 ...
## $ education: Factor w/ 4 levels "primary","secondary",..: 1 2 3 3 2 3 3 2 3 1 ...
## $ default : Factor w/ 2 levels "no","yes": 1 1 1 1 1 1 1 1 1 1 ...
## $ balance : num [1:8000] 1787 4789 1350 1476 0 ...
## $ housing : Factor w/ 2 levels "no","yes": 1 2 2 2 2 1 2 2 2 2 ...
## $ loan : Factor w/ 2 levels "no","yes": 1 2 1 2 1 1 1 1 1 2 ...
## $ contact : Factor w/ 3 levels "cellular","telephone",..: 1 1 1 3 3 1 1 1 3 1 ...
## $ day : num [1:8000] 19 11 16 3 5 23 14 6 14 17 ...
## $ month : Factor w/ 12 levels "apr","aug","dec",..: 11 9 1 7 9 4 9 9 9 1 ...
## $ duration : num [1:8000] 79 220 185 199 226 141 341 151 57 313 ...
## $ campaign : num [1:8000] 1 1 1 4 1 2 1 2 2 1 ...
## $ pdays : num [1:8000] -1 339 330 -1 -1 176 330 -1 -1 147 ...
## $ previous : num [1:8000] 0 4 1 0 0 3 2 0 0 2 ...
## $ poutcome : Factor w/ 4 levels "failure","other",..: 4 1 1 4 4 1 2 4 4 1 ...
## $ y : Factor w/ 2 levels "no","yes": 1 1 1 1 1 1 1 1 1 1 ...# Coba ANN
# OHE untuk seluruh variabel kategorik
categorical_cols <- c("job", "marital", "education", "default", "housing", "loan", "contact", "month", "poutcome")
# Gabungkan semua hasil OHE dari semua kolom
ohe_all <- do.call(cbind, lapply(categorical_cols, function(col) {
model.matrix(~ . -1, data = data[col])
}))
data_numeric <- data[ , !(names(data) %in% categorical_cols)]
data2 <- cbind(data_numeric, ohe_all)
head(data2)set.seed(123)
train.index <- createDataPartition(data2$y, p = 0.8, list = FALSE)
train2 <- data2[train.index, ]
test2 <- data2[-train.index, ]
# Preprocessing untuk scaling range (min-max ke 0-1)
preprocessParams <- preProcess(train2[, -8], method = c("range"))
# Transformasi data training dan testing
train_X <- as.matrix(predict(preprocessParams, train2[, -8]))
test_X <- as.matrix(predict(preprocessParams, test2[, -8]))
train_y <- as.numeric(train2[,8])-1
test_y <- as.numeric(test2[,8])-1
# Gabungkan data fitur dan target menjadi satu data frame
train_data <- data.frame(train_X)
train_data$y <- train_y
test_data <- data.frame(test_X)
test_data$y <- test_y
head(train_data)head(test_data)# Neural Net Model
n <- names(train_data)
f <- as.formula(paste("y ~", paste(n[!n %in% "y"], collapse = " + ")))
nn <- neuralnet(f, data=train_data, hidden=c(8,5,3), linear.output=FALSE)Prediksi dan evaluasi data training
# hitung prediksi
pr.nn <- compute(nn, train_data)
# Klasifikasi biner (threshold 0.5)
roundedresults <- ifelse(pr.nn$net.result > 0.5, 1, 0)
# Gabungkan dengan nilai aktual
results <- data.frame(actual = train_data$y, prediction = roundedresults)
# Ubah ke factor dengan level yang sama dan urutan sama
actual <- factor(results$actual, levels = c(0, 1))
predicted <- factor(results$prediction, levels = c(0, 1))
library(caret)
library(MLmetrics)
# Confusion Matrix
conf_mat <- confusionMatrix(predicted, actual, positive = "0")
print(conf_mat)## Confusion Matrix and Statistics
##
## Reference
## Prediction 0 1
## 0 3036 104
## 1 164 3096
##
## Accuracy : 0.9581
## 95% CI : (0.9529, 0.9629)
## No Information Rate : 0.5
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.9162
##
## Mcnemar's Test P-Value : 0.0003134
##
## Sensitivity : 0.9487
## Specificity : 0.9675
## Pos Pred Value : 0.9669
## Neg Pred Value : 0.9497
## Prevalence : 0.5000
## Detection Rate : 0.4744
## Detection Prevalence : 0.4906
## Balanced Accuracy : 0.9581
##
## 'Positive' Class : 0
## precision <- conf_mat$byClass["Precision"]
recall <- conf_mat$byClass["Recall"]
f1 <- 2 * ((precision * recall) / (precision + recall))
cat("Precision:", round(precision, 4), "\n")## Precision: 0.9669cat("Recall :", round(recall, 4), "\n")## Recall : 0.9488cat("F1 Score :", round(f1, 4), "\n")## F1 Score : 0.9577Prediksi dan evaluasi data testing
# hitung prediksi
pr.nn <- compute(nn, test_data)
# Klasifikasi biner (threshold 0.5)
roundedresults <- ifelse(pr.nn$net.result > 0.5, 1, 0)
# Gabungkan dengan nilai aktual
results <- data.frame(actual = test_data$y, prediction = roundedresults)
# Ubah ke factor dengan level yang sama dan urutan sama
actual <- factor(results$actual, levels = c(0, 1))
predicted <- factor(results$prediction, levels = c(0, 1))
library(caret)
library(MLmetrics)
# Confusion Matrix
conf_mat <- confusionMatrix(predicted, actual, positive = "0")
print(conf_mat)## Confusion Matrix and Statistics
##
## Reference
## Prediction 0 1
## 0 702 94
## 1 98 706
##
## Accuracy : 0.88
## 95% CI : (0.8631, 0.8955)
## No Information Rate : 0.5
## P-Value [Acc > NIR] : <2e-16
##
## Kappa : 0.76
##
## Mcnemar's Test P-Value : 0.8286
##
## Sensitivity : 0.8775
## Specificity : 0.8825
## Pos Pred Value : 0.8819
## Neg Pred Value : 0.8781
## Prevalence : 0.5000
## Detection Rate : 0.4387
## Detection Prevalence : 0.4975
## Balanced Accuracy : 0.8800
##
## 'Positive' Class : 0
## precision <- conf_mat$byClass["Precision"]
recall <- conf_mat$byClass["Recall"]
f1 <- 2 * ((precision * recall) / (precision + recall))
cat("Precision:", round(precision, 4), "\n")## Precision: 0.8819cat("Recall :", round(recall, 4), "\n")## Recall : 0.8775cat("F1 Score :", round(f1, 4), "\n")## F1 Score : 0.8797Visualisasi model ANN
nn$result.matrix## [,1]
## error 1.340071e+02
## reached.threshold 9.706640e-03
## steps 3.613100e+04
## Intercept.to.1layhid1 1.975204e+00
## Age.to.1layhid1 1.668632e+00
## balance.to.1layhid1 -3.040156e+01
## day.to.1layhid1 -9.076192e-01
## duration.to.1layhid1 -2.957858e+00
## campaign.to.1layhid1 -3.683431e+00
## pdays.to.1layhid1 -3.753081e+00
## previous.to.1layhid1 -5.223280e+01
## jobadmin..to.1layhid1 1.535609e+00
## jobblue.collar.to.1layhid1 2.714064e+00
## jobentrepreneur.to.1layhid1 -6.463774e+00
## jobhousemaid.to.1layhid1 4.326269e+00
## jobmanagement.to.1layhid1 4.925302e-01
## jobretired.to.1layhid1 1.301833e+00
## jobself.employed.to.1layhid1 5.464124e+00
## jobservices.to.1layhid1 4.012648e+00
## jobstudent.to.1layhid1 5.170893e+00
## jobtechnician.to.1layhid1 -3.925145e-01
## jobunemployed.to.1layhid1 -1.522507e+00
## jobunknown.to.1layhid1 8.810766e-03
## maritaldivorced.to.1layhid1 2.896108e+00
## maritalmarried.to.1layhid1 -1.914043e-01
## maritalsingle.to.1layhid1 2.031094e+00
## educationprimary.to.1layhid1 4.688999e+00
## educationsecondary.to.1layhid1 -7.208166e-01
## educationtertiary.to.1layhid1 -6.530851e-01
## educationunknown.to.1layhid1 1.027968e+01
## defaultno.to.1layhid1 -4.925499e-01
## defaultyes.to.1layhid1 7.521522e+01
## housingno.to.1layhid1 4.447278e-01
## housingyes.to.1layhid1 -5.698732e-01
## loanno.to.1layhid1 -3.513318e-01
## loanyes.to.1layhid1 1.275892e+00
## contactcellular.to.1layhid1 -1.698724e+00
## contacttelephone.to.1layhid1 4.203408e+01
## contactunknown.to.1layhid1 9.108666e+01
## monthapr.to.1layhid1 -9.285824e+00
## monthaug.to.1layhid1 8.338260e-01
## monthdec.to.1layhid1 -7.407754e+02
## monthfeb.to.1layhid1 -3.464521e+00
## monthjan.to.1layhid1 -1.046900e+00
## monthjul.to.1layhid1 1.058604e+01
## monthjun.to.1layhid1 -4.136037e+01
## monthmar.to.1layhid1 1.987118e+00
## monthmay.to.1layhid1 2.534901e+00
## monthnov.to.1layhid1 1.934157e+00
## monthoct.to.1layhid1 -8.667004e+00
## monthsep.to.1layhid1 -1.342122e+00
## poutcomefailure.to.1layhid1 -8.762008e-01
## poutcomeother.to.1layhid1 -2.474319e+01
## poutcomesuccess.to.1layhid1 -8.732492e+00
## poutcomeunknown.to.1layhid1 7.187486e-01
## Intercept.to.1layhid2 1.638201e+00
## Age.to.1layhid2 -1.083790e+01
## balance.to.1layhid2 2.854567e+00
## day.to.1layhid2 2.518339e-01
## duration.to.1layhid2 -1.482113e+01
## campaign.to.1layhid2 6.171783e+00
## pdays.to.1layhid2 -8.574658e+01
## previous.to.1layhid2 -8.292213e+01
## jobadmin..to.1layhid2 -2.451990e+00
## jobblue.collar.to.1layhid2 8.280909e-01
## jobentrepreneur.to.1layhid2 -4.487023e+00
## jobhousemaid.to.1layhid2 2.116652e+01
## jobmanagement.to.1layhid2 2.167646e+00
## jobretired.to.1layhid2 -7.351176e+02
## jobself.employed.to.1layhid2 -7.113095e-01
## jobservices.to.1layhid2 6.003828e+01
## jobstudent.to.1layhid2 2.898153e+00
## jobtechnician.to.1layhid2 -1.296873e+00
## jobunemployed.to.1layhid2 1.994518e+01
## jobunknown.to.1layhid2 4.678846e+00
## maritaldivorced.to.1layhid2 -4.702801e-02
## maritalmarried.to.1layhid2 2.965980e-01
## maritalsingle.to.1layhid2 1.219299e+00
## educationprimary.to.1layhid2 -1.257842e+01
## educationsecondary.to.1layhid2 -1.205702e-01
## educationtertiary.to.1layhid2 -1.435531e+00
## educationunknown.to.1layhid2 1.893297e+00
## defaultno.to.1layhid2 1.822150e+00
## defaultyes.to.1layhid2 9.343637e+00
## housingno.to.1layhid2 5.748477e-01
## housingyes.to.1layhid2 2.560908e-03
## loanno.to.1layhid2 -3.708647e-01
## loanyes.to.1layhid2 -7.261535e-01
## contactcellular.to.1layhid2 -1.170709e+00
## contacttelephone.to.1layhid2 -8.285774e+00
## contactunknown.to.1layhid2 8.494272e+00
## monthapr.to.1layhid2 -2.495126e+00
## monthaug.to.1layhid2 3.968568e+00
## monthdec.to.1layhid2 -1.437963e+01
## monthfeb.to.1layhid2 -1.643491e+00
## monthjan.to.1layhid2 6.603103e+00
## monthjul.to.1layhid2 3.488060e+00
## monthjun.to.1layhid2 -4.124838e-02
## monthmar.to.1layhid2 -3.401314e+00
## monthmay.to.1layhid2 3.769992e+00
## monthnov.to.1layhid2 -6.255529e+01
## monthoct.to.1layhid2 2.354466e+00
## monthsep.to.1layhid2 4.220002e+00
## poutcomefailure.to.1layhid2 -9.954650e+00
## poutcomeother.to.1layhid2 -4.273616e+01
## poutcomesuccess.to.1layhid2 2.688975e+01
## poutcomeunknown.to.1layhid2 6.788900e-01
## Intercept.to.1layhid3 -1.211151e-01
## Age.to.1layhid3 3.767028e+00
## balance.to.1layhid3 -5.717472e+00
## day.to.1layhid3 1.859613e+00
## duration.to.1layhid3 7.529707e+00
## campaign.to.1layhid3 -4.909695e+00
## pdays.to.1layhid3 -3.458005e+00
## previous.to.1layhid3 3.071345e+00
## jobadmin..to.1layhid3 2.406757e+00
## jobblue.collar.to.1layhid3 -1.552772e+00
## jobentrepreneur.to.1layhid3 -3.639860e-01
## jobhousemaid.to.1layhid3 3.735712e+01
## jobmanagement.to.1layhid3 1.689742e+00
## jobretired.to.1layhid3 -2.676279e-01
## jobself.employed.to.1layhid3 -4.779314e-01
## jobservices.to.1layhid3 2.895380e+00
## jobstudent.to.1layhid3 4.317940e+00
## jobtechnician.to.1layhid3 3.785529e-01
## jobunemployed.to.1layhid3 2.182154e+00
## jobunknown.to.1layhid3 3.839135e+01
## maritaldivorced.to.1layhid3 4.478692e+00
## maritalmarried.to.1layhid3 -2.354050e-01
## maritalsingle.to.1layhid3 -1.167574e+00
## educationprimary.to.1layhid3 -2.702746e-02
## educationsecondary.to.1layhid3 1.938854e+00
## educationtertiary.to.1layhid3 1.303102e+00
## educationunknown.to.1layhid3 -1.568601e+00
## defaultno.to.1layhid3 -9.463171e-02
## defaultyes.to.1layhid3 -3.119093e+00
## housingno.to.1layhid3 1.224395e+00
## housingyes.to.1layhid3 7.633154e-01
## loanno.to.1layhid3 -4.339480e-01
## loanyes.to.1layhid3 -1.555969e+00
## contactcellular.to.1layhid3 -1.085216e+00
## contacttelephone.to.1layhid3 2.426987e+00
## contactunknown.to.1layhid3 1.264403e+00
## monthapr.to.1layhid3 -3.009921e+00
## monthaug.to.1layhid3 2.447327e+00
## monthdec.to.1layhid3 -1.532837e+00
## monthfeb.to.1layhid3 -8.133975e-01
## monthjan.to.1layhid3 5.608743e-01
## monthjul.to.1layhid3 -3.803332e-02
## monthjun.to.1layhid3 -7.907242e-01
## monthmar.to.1layhid3 -6.112525e-01
## monthmay.to.1layhid3 -6.092909e-01
## monthnov.to.1layhid3 2.299170e-01
## monthoct.to.1layhid3 -1.507089e+00
## monthsep.to.1layhid3 1.936756e+00
## poutcomefailure.to.1layhid3 -1.373468e+00
## poutcomeother.to.1layhid3 -4.362537e+00
## poutcomesuccess.to.1layhid3 2.950444e-01
## poutcomeunknown.to.1layhid3 -2.960709e-03
## Intercept.to.1layhid4 -2.013953e-02
## Age.to.1layhid4 1.874468e+00
## balance.to.1layhid4 -1.121124e+01
## day.to.1layhid4 -8.087830e-01
## duration.to.1layhid4 2.582736e+00
## campaign.to.1layhid4 -5.269388e+00
## pdays.to.1layhid4 1.633003e+01
## previous.to.1layhid4 6.736636e+00
## jobadmin..to.1layhid4 3.082387e+00
## jobblue.collar.to.1layhid4 6.833489e-01
## jobentrepreneur.to.1layhid4 6.389576e+00
## jobhousemaid.to.1layhid4 3.090919e+00
## jobmanagement.to.1layhid4 -1.044484e+00
## jobretired.to.1layhid4 2.466396e+00
## jobself.employed.to.1layhid4 6.204782e+00
## jobservices.to.1layhid4 3.377038e-01
## jobstudent.to.1layhid4 2.059337e+00
## jobtechnician.to.1layhid4 -1.637070e+00
## jobunemployed.to.1layhid4 3.223043e-01
## jobunknown.to.1layhid4 3.387101e+00
## maritaldivorced.to.1layhid4 -2.265562e+00
## maritalmarried.to.1layhid4 -5.437309e-02
## maritalsingle.to.1layhid4 1.256398e+00
## educationprimary.to.1layhid4 -3.338930e+00
## educationsecondary.to.1layhid4 1.146023e+00
## educationtertiary.to.1layhid4 -2.094604e-01
## educationunknown.to.1layhid4 -2.572655e+00
## defaultno.to.1layhid4 -1.847496e+00
## defaultyes.to.1layhid4 3.758072e+00
## housingno.to.1layhid4 1.141656e+00
## housingyes.to.1layhid4 -1.261458e+00
## loanno.to.1layhid4 -1.128927e+00
## loanyes.to.1layhid4 3.971313e+01
## contactcellular.to.1layhid4 1.813991e+00
## contacttelephone.to.1layhid4 -3.848781e+00
## contactunknown.to.1layhid4 2.400305e-01
## monthapr.to.1layhid4 1.297694e+00
## monthaug.to.1layhid4 -3.838629e+00
## monthdec.to.1layhid4 4.462433e+01
## monthfeb.to.1layhid4 3.048197e+00
## monthjan.to.1layhid4 1.778936e+00
## monthjul.to.1layhid4 2.166718e+00
## monthjun.to.1layhid4 -1.851284e+00
## monthmar.to.1layhid4 -3.120169e-02
## monthmay.to.1layhid4 1.974647e-01
## monthnov.to.1layhid4 2.127073e-01
## monthoct.to.1layhid4 3.217641e+00
## monthsep.to.1layhid4 4.512883e+01
## poutcomefailure.to.1layhid4 -1.059641e+00
## poutcomeother.to.1layhid4 -6.085642e+00
## poutcomesuccess.to.1layhid4 1.049240e+00
## poutcomeunknown.to.1layhid4 1.150864e+00
## Intercept.to.1layhid5 -8.014707e-01
## Age.to.1layhid5 2.118969e+00
## balance.to.1layhid5 5.886851e+00
## day.to.1layhid5 1.367999e-01
## duration.to.1layhid5 -7.280472e+00
## campaign.to.1layhid5 -1.962592e+00
## pdays.to.1layhid5 -1.152373e+00
## previous.to.1layhid5 4.318907e+00
## jobadmin..to.1layhid5 1.095796e+00
## jobblue.collar.to.1layhid5 4.233552e-01
## jobentrepreneur.to.1layhid5 1.133980e+00
## jobhousemaid.to.1layhid5 -2.052282e+00
## jobmanagement.to.1layhid5 -2.967160e-01
## jobretired.to.1layhid5 -1.182918e-01
## jobself.employed.to.1layhid5 1.892145e+00
## jobservices.to.1layhid5 1.224944e-01
## jobstudent.to.1layhid5 -5.981062e-01
## jobtechnician.to.1layhid5 -8.793426e-02
## jobunemployed.to.1layhid5 -1.068323e+00
## jobunknown.to.1layhid5 -7.784582e-01
## maritaldivorced.to.1layhid5 8.096876e-01
## maritalmarried.to.1layhid5 9.197937e-01
## maritalsingle.to.1layhid5 5.241361e-01
## educationprimary.to.1layhid5 -6.193330e-01
## educationsecondary.to.1layhid5 -1.439214e-01
## educationtertiary.to.1layhid5 -2.263765e+00
## educationunknown.to.1layhid5 -4.674116e+00
## defaultno.to.1layhid5 1.332003e+00
## defaultyes.to.1layhid5 1.873611e-01
## housingno.to.1layhid5 7.509898e-01
## housingyes.to.1layhid5 4.656725e-01
## loanno.to.1layhid5 -4.357132e-01
## loanyes.to.1layhid5 1.704860e+00
## contactcellular.to.1layhid5 -2.179870e+00
## contacttelephone.to.1layhid5 -1.546849e+00
## contactunknown.to.1layhid5 -1.632913e+00
## monthapr.to.1layhid5 1.135625e+00
## monthaug.to.1layhid5 -4.044784e-02
## monthdec.to.1layhid5 1.142537e+00
## monthfeb.to.1layhid5 -9.246845e-01
## monthjan.to.1layhid5 -8.148539e-01
## monthjul.to.1layhid5 8.254016e-01
## monthjun.to.1layhid5 7.177342e-01
## monthmar.to.1layhid5 -9.880548e-01
## monthmay.to.1layhid5 -8.768715e-02
## monthnov.to.1layhid5 3.491425e-01
## monthoct.to.1layhid5 -9.517497e-02
## monthsep.to.1layhid5 1.727633e+00
## poutcomefailure.to.1layhid5 -8.598706e-01
## poutcomeother.to.1layhid5 -9.158963e-02
## poutcomesuccess.to.1layhid5 5.884042e-01
## poutcomeunknown.to.1layhid5 -1.176170e-01
## Intercept.to.1layhid6 3.240919e-01
## Age.to.1layhid6 6.227236e-01
## balance.to.1layhid6 -1.366494e+00
## day.to.1layhid6 7.583054e-01
## duration.to.1layhid6 -6.547001e+00
## campaign.to.1layhid6 7.076069e+00
## pdays.to.1layhid6 -4.169861e+00
## previous.to.1layhid6 4.116486e+00
## jobadmin..to.1layhid6 -2.605494e+00
## jobblue.collar.to.1layhid6 -1.019714e+00
## jobentrepreneur.to.1layhid6 6.581063e-01
## jobhousemaid.to.1layhid6 1.673953e+00
## jobmanagement.to.1layhid6 -8.015288e-01
## jobretired.to.1layhid6 1.431530e+00
## jobself.employed.to.1layhid6 -2.300897e+00
## jobservices.to.1layhid6 -2.908955e+00
## jobstudent.to.1layhid6 -7.834468e-01
## jobtechnician.to.1layhid6 -1.995835e+00
## jobunemployed.to.1layhid6 3.906728e-01
## jobunknown.to.1layhid6 3.993257e+00
## maritaldivorced.to.1layhid6 -4.724340e-01
## maritalmarried.to.1layhid6 -9.935489e-01
## maritalsingle.to.1layhid6 -8.932758e-01
## educationprimary.to.1layhid6 1.545002e+00
## educationsecondary.to.1layhid6 6.237506e-01
## educationtertiary.to.1layhid6 1.735870e+00
## educationunknown.to.1layhid6 9.323191e-01
## defaultno.to.1layhid6 -7.406047e-01
## defaultyes.to.1layhid6 -2.273366e+00
## housingno.to.1layhid6 5.158255e-01
## housingyes.to.1layhid6 5.485593e-01
## loanno.to.1layhid6 1.518058e+00
## loanyes.to.1layhid6 5.847989e-01
## contactcellular.to.1layhid6 -1.818752e+00
## contacttelephone.to.1layhid6 -5.547683e+00
## contactunknown.to.1layhid6 -2.218675e+00
## monthapr.to.1layhid6 -1.889535e+00
## monthaug.to.1layhid6 1.043774e-01
## monthdec.to.1layhid6 -2.948089e+00
## monthfeb.to.1layhid6 3.599427e+00
## monthjan.to.1layhid6 7.610154e-02
## monthjul.to.1layhid6 -2.843504e-02
## monthjun.to.1layhid6 8.953096e-02
## monthmar.to.1layhid6 1.038799e+00
## monthmay.to.1layhid6 -1.687799e-01
## monthnov.to.1layhid6 9.113074e-01
## monthoct.to.1layhid6 1.014731e+00
## monthsep.to.1layhid6 -2.428020e+00
## poutcomefailure.to.1layhid6 1.469164e+00
## poutcomeother.to.1layhid6 -2.192679e+00
## poutcomesuccess.to.1layhid6 -1.687175e+00
## poutcomeunknown.to.1layhid6 -4.228287e-01
## Intercept.to.1layhid7 2.026868e-03
## Age.to.1layhid7 -1.448299e-01
## balance.to.1layhid7 1.768417e+00
## day.to.1layhid7 -3.983240e-01
## duration.to.1layhid7 -6.443581e+00
## campaign.to.1layhid7 1.058362e+01
## pdays.to.1layhid7 -1.382957e+01
## previous.to.1layhid7 -2.310302e+01
## jobadmin..to.1layhid7 1.060357e+00
## jobblue.collar.to.1layhid7 -9.854690e-01
## jobentrepreneur.to.1layhid7 7.143458e-01
## jobhousemaid.to.1layhid7 -1.436132e+00
## jobmanagement.to.1layhid7 -1.423057e+00
## jobretired.to.1layhid7 -5.802621e+00
## jobself.employed.to.1layhid7 -2.194979e-01
## jobservices.to.1layhid7 7.175454e-01
## jobstudent.to.1layhid7 -2.142418e+00
## jobtechnician.to.1layhid7 8.701030e-01
## jobunemployed.to.1layhid7 -1.543054e+00
## jobunknown.to.1layhid7 -9.775145e-01
## maritaldivorced.to.1layhid7 -1.156470e+00
## maritalmarried.to.1layhid7 -6.432414e-02
## maritalsingle.to.1layhid7 -6.119126e-01
## educationprimary.to.1layhid7 -2.693771e+00
## educationsecondary.to.1layhid7 8.062236e-02
## educationtertiary.to.1layhid7 1.819977e+00
## educationunknown.to.1layhid7 -3.702648e+00
## defaultno.to.1layhid7 -8.849730e-01
## defaultyes.to.1layhid7 3.780477e+00
## housingno.to.1layhid7 -2.077743e-01
## housingyes.to.1layhid7 8.940361e-01
## loanno.to.1layhid7 -4.715617e-01
## loanyes.to.1layhid7 6.182554e+00
## contactcellular.to.1layhid7 -8.275624e-01
## contacttelephone.to.1layhid7 -1.309059e+00
## contactunknown.to.1layhid7 -5.532929e+00
## monthapr.to.1layhid7 1.227994e+00
## monthaug.to.1layhid7 -3.261886e-01
## monthdec.to.1layhid7 1.602546e+00
## monthfeb.to.1layhid7 4.205807e-01
## monthjan.to.1layhid7 7.420173e+00
## monthjul.to.1layhid7 5.846033e-02
## monthjun.to.1layhid7 -1.469159e-01
## monthmar.to.1layhid7 -7.122293e+02
## monthmay.to.1layhid7 -6.531647e-01
## monthnov.to.1layhid7 2.412804e+00
## monthoct.to.1layhid7 1.611539e+00
## monthsep.to.1layhid7 -1.024100e+01
## poutcomefailure.to.1layhid7 3.370180e+00
## poutcomeother.to.1layhid7 -7.155617e+02
## poutcomesuccess.to.1layhid7 -3.526581e-01
## poutcomeunknown.to.1layhid7 1.441394e+00
## Intercept.to.1layhid8 -1.137443e+00
## Age.to.1layhid8 -8.046675e-01
## balance.to.1layhid8 2.185474e+00
## day.to.1layhid8 -1.056169e+00
## duration.to.1layhid8 -5.607305e+00
## campaign.to.1layhid8 -1.106516e+01
## pdays.to.1layhid8 1.747275e+00
## previous.to.1layhid8 -3.256033e-01
## jobadmin..to.1layhid8 -1.368639e-01
## jobblue.collar.to.1layhid8 -1.555564e+00
## jobentrepreneur.to.1layhid8 -2.340250e+00
## jobhousemaid.to.1layhid8 -1.256247e+00
## jobmanagement.to.1layhid8 -1.013798e+00
## jobretired.to.1layhid8 -7.351182e-01
## jobself.employed.to.1layhid8 -1.516385e-01
## jobservices.to.1layhid8 -7.202693e-02
## jobstudent.to.1layhid8 1.756603e+00
## jobtechnician.to.1layhid8 -9.075744e-01
## jobunemployed.to.1layhid8 1.763425e+00
## jobunknown.to.1layhid8 -7.087506e+02
## maritaldivorced.to.1layhid8 6.280788e-01
## maritalmarried.to.1layhid8 -5.342433e-02
## maritalsingle.to.1layhid8 3.179200e-01
## educationprimary.to.1layhid8 -2.118051e+00
## educationsecondary.to.1layhid8 -8.975179e-03
## educationtertiary.to.1layhid8 5.234190e-01
## educationunknown.to.1layhid8 5.753391e+00
## defaultno.to.1layhid8 -2.182268e-01
## defaultyes.to.1layhid8 3.438158e-01
## housingno.to.1layhid8 -3.834396e-01
## housingyes.to.1layhid8 4.210381e-02
## loanno.to.1layhid8 2.263096e-01
## loanyes.to.1layhid8 -1.139635e+00
## contactcellular.to.1layhid8 -1.044397e+00
## contacttelephone.to.1layhid8 6.537770e-01
## contactunknown.to.1layhid8 1.204741e-01
## monthapr.to.1layhid8 -8.274409e-02
## monthaug.to.1layhid8 -1.102287e+00
## monthdec.to.1layhid8 4.441289e+00
## monthfeb.to.1layhid8 -1.954711e-01
## monthjan.to.1layhid8 1.403906e+00
## monthjul.to.1layhid8 -1.992369e+00
## monthjun.to.1layhid8 -2.503076e+00
## monthmar.to.1layhid8 7.776419e-02
## monthmay.to.1layhid8 1.238685e+00
## monthnov.to.1layhid8 -6.105735e-01
## monthoct.to.1layhid8 -2.279546e+00
## monthsep.to.1layhid8 2.026464e+00
## poutcomefailure.to.1layhid8 3.129039e+00
## poutcomeother.to.1layhid8 2.787396e+00
## poutcomesuccess.to.1layhid8 -3.621444e+00
## poutcomeunknown.to.1layhid8 -1.663585e-02
## Intercept.to.2layhid1 -5.858536e-01
## 1layhid1.to.2layhid1 -9.199081e-02
## 1layhid2.to.2layhid1 1.092584e+00
## 1layhid3.to.2layhid1 -1.923186e+00
## 1layhid4.to.2layhid1 -5.520406e-01
## 1layhid5.to.2layhid1 2.616454e+00
## 1layhid6.to.2layhid1 1.471471e+00
## 1layhid7.to.2layhid1 1.822767e+00
## 1layhid8.to.2layhid1 2.370176e+00
## Intercept.to.2layhid2 5.394650e-01
## 1layhid1.to.2layhid2 -4.661477e-02
## 1layhid2.to.2layhid2 -1.324330e-01
## 1layhid3.to.2layhid2 -4.041279e-01
## 1layhid4.to.2layhid2 1.306686e+00
## 1layhid5.to.2layhid2 -2.376670e+00
## 1layhid6.to.2layhid2 -2.462045e+00
## 1layhid7.to.2layhid2 1.403624e-01
## 1layhid8.to.2layhid2 -2.754111e+00
## Intercept.to.2layhid3 1.169173e-01
## 1layhid1.to.2layhid3 -7.146576e-01
## 1layhid2.to.2layhid3 -3.201807e-01
## 1layhid3.to.2layhid3 1.400138e+00
## 1layhid4.to.2layhid3 1.187377e+00
## 1layhid5.to.2layhid3 -1.752048e+00
## 1layhid6.to.2layhid3 -2.741954e+00
## 1layhid7.to.2layhid3 -1.210214e+00
## 1layhid8.to.2layhid3 -2.889511e+00
## Intercept.to.2layhid4 4.904340e-01
## 1layhid1.to.2layhid4 1.822800e+00
## 1layhid2.to.2layhid4 1.737800e+00
## 1layhid3.to.2layhid4 -2.602749e+00
## 1layhid4.to.2layhid4 -1.410385e+00
## 1layhid5.to.2layhid4 1.372051e+00
## 1layhid6.to.2layhid4 1.877793e+00
## 1layhid7.to.2layhid4 1.399991e+00
## 1layhid8.to.2layhid4 3.022029e+00
## Intercept.to.2layhid5 1.174216e+00
## 1layhid1.to.2layhid5 -1.477989e+00
## 1layhid2.to.2layhid5 5.228013e-01
## 1layhid3.to.2layhid5 1.170513e-01
## 1layhid4.to.2layhid5 1.318623e+00
## 1layhid5.to.2layhid5 -1.921424e+00
## 1layhid6.to.2layhid5 -1.153981e+00
## 1layhid7.to.2layhid5 -2.025604e+00
## 1layhid8.to.2layhid5 -2.511778e+00
## Intercept.to.3layhid1 1.405934e+00
## 2layhid1.to.3layhid1 2.361183e+00
## 2layhid2.to.3layhid1 -3.615932e-01
## 2layhid3.to.3layhid1 -7.873079e-01
## 2layhid4.to.3layhid1 1.263847e+00
## 2layhid5.to.3layhid1 -3.403719e+00
## Intercept.to.3layhid2 1.318123e+00
## 2layhid1.to.3layhid2 -2.739770e+00
## 2layhid2.to.3layhid2 5.292493e-01
## 2layhid3.to.3layhid2 6.552461e-01
## 2layhid4.to.3layhid2 -3.261291e+00
## 2layhid5.to.3layhid2 1.468879e+00
## Intercept.to.3layhid3 -1.506847e+00
## 2layhid1.to.3layhid3 -4.517012e+00
## 2layhid2.to.3layhid3 7.341205e-01
## 2layhid3.to.3layhid3 2.616834e+00
## 2layhid4.to.3layhid3 -1.235771e-01
## 2layhid5.to.3layhid3 2.275952e+00
## Intercept.to.y -1.026662e+00
## 3layhid1.to.y -2.169668e+02
## 3layhid2.to.y 4.776645e+02
## 3layhid3.to.y 6.787929e+01summary(nn)## Length Class Mode
## call 5 -none- call
## response 6400 -none- numeric
## covariate 326400 -none- numeric
## model.list 2 -none- list
## err.fct 1 -none- function
## act.fct 1 -none- function
## linear.output 1 -none- logical
## data 52 data.frame list
## exclude 0 -none- NULL
## net.result 1 -none- list
## weights 1 -none- list
## generalized.weights 1 -none- list
## startweights 1 -none- list
## result.matrix 486 -none- numericplot(nn)