Classifications
Arwan Zhagi
4/13/2021
Visualisasi Data
Memuat Package
library(tidyverse)
library(ggplot2)Dataset
datairis <-
read.csv("Dataset Iris.csv", sep = ";")
submission <-
read.csv("Data iris submission.csv", sep = ";")Misssing Data
datairis %>%
count(is.na(.))## is.na(.).Id is.na(.).SepalLengthCm is.na(.).SepalWidthCm
## 1 FALSE FALSE FALSE
## 2 FALSE FALSE FALSE
## 3 FALSE FALSE FALSE
## 4 FALSE FALSE TRUE
## 5 FALSE TRUE FALSE
## is.na(.).PetalLengthCm is.na(.).PetalWidthCm is.na(.).Species n
## 1 FALSE FALSE FALSE 122
## 2 FALSE TRUE FALSE 3
## 3 TRUE FALSE FALSE 2
## 4 FALSE FALSE FALSE 3
## 5 FALSE FALSE FALSE 5datairis <-
datairis %>%
drop_na()Distribusi Data Berdasarkan jenis Bunga
datairis %>%
ggplot() +
geom_boxplot(aes(Species, SepalWidthCm))+
labs(title = "Distribusi Data berdasarkan Jenis Bunga",
subtitle = "Sepal Width Cm",
caption = "by : Arwan Zhagi")+
xlab("Jenis Bunga")+
ylab("Sepal Width Cm")+
theme(plot.title = element_text(color = "blue",size = 17, face = "bold", hjust = 0.5),
plot.subtitle = element_text(size = 10, face = "bold"))
datairis %>%
ggplot() +
geom_boxplot(aes(Species, SepalLengthCm))+
labs(title = "Distribusi Data berdasarkan Jenis Bunga",
subtitle = "Sepal Length Cm",
caption = "by : Arwan Zhagi")+
xlab("Jenis Bunga")+
ylab("Sepal Length Cm")+
theme(plot.title = element_text(color = "blue",size = 17, face = "bold", hjust = 0.5),
plot.subtitle = element_text(size = 10, face = "bold"))
datairis %>%
ggplot() +
geom_boxplot(aes(Species, PetalWidthCm))+
labs(title = "Distribusi Data berdasarkan Jenis Bunga",
subtitle = "Petal Width Cm",
caption = "by : Arwan Zhagi")+
xlab("Jenis Bunga")+
ylab("Petal Width Cm")+
theme(plot.title = element_text(color = "blue",size = 17, face = "bold", hjust = 0.5),
plot.subtitle = element_text(size = 10, face = "bold"))
datairis %>%
ggplot() +
geom_boxplot(aes(Species, PetalLengthCm), show.legend = T)+
labs(title = "Distribusi Data berdasarkan Jenis Bunga",
subtitle = "Petal Length Cm",
caption = "by : Arwan Zhagi")+
xlab("Jenis Bunga")+
ylab("Petal Length Cm")+
theme(plot.title = element_text(color = "blue",size = 17, face = "bold", hjust = 0.5),
plot.subtitle = element_text(size = 10, face = "bold"))
Hubungan Antara Dua Variabel Berdaasarkan Jenis Bunga
datairis%>%
ggplot(aes(SepalWidthCm,SepalLengthCm))+
geom_point(aes(color = Species))+
facet_grid(. ~ Species)+
labs(title = "Hubungan Sepal Length dengan\nSepal Width Berdasarkan\nJenis Bunga ",
caption = "by : Arwan Zhagi")+
xlab("Sepal Length")+
ylab("Sepal Width")+
theme(plot.title = element_text(color = "blue",size = 17, face = "bold", hjust = 0.5),
)
datairis%>%
ggplot(aes(SepalWidthCm,PetalLengthCm))+
geom_point(aes(color = Species))+
facet_grid(. ~ Species)+
labs(title = "Hubungan Sepal Width dengan\npetal Length Berdasarkan\nJenis Bunga ",
caption = "by : Arwan Zhagi")+
xlab("Sepal Width")+
ylab("petal Length")+
theme(plot.title = element_text(color = "blue",size = 17, face = "bold", hjust = 0.5),
)
datairis%>%
ggplot(aes(PetalWidthCm,SepalLengthCm))+
geom_point(aes(color = Species))+
facet_grid(. ~ Species)+
labs(title = "Hubungan Petal Width dengan\nSepal Length Berdasarkan\nJenis Bunga ",
caption = "by : Arwan Zhagi")+
xlab("Petal Width")+
ylab("Sepal Length")+
theme(plot.title = element_text(color = "blue",size = 17, face = "bold", hjust = 0.5),
)
datairis%>%
ggplot(aes(PetalWidthCm,PetalLengthCm))+
geom_point(aes(color = Species))+
facet_grid(. ~ Species)+
labs(title = "Hubungan Petal Width dengan\nPetal Length Berdasarkan\nJenis Bunga ",
caption = "by : Arwan Zhagi")+
xlab("Petal Width")+
ylab("Petal Length")+
theme(plot.title = element_text(color = "blue",size = 17, face = "bold", hjust = 0.5),
)
CLASSIFICATIONS
Dataset Species Diubah Menjadi Numerik
datairis <-
datairis %>%
mutate(SpeciesNum =
ifelse(Species == "Iris-setosa",1,
ifelse(Species == "Iris-versicolor",2,3)))Regresi Logistik
Memuat Package
library(tidyverse)
library(caret)Model Dataset Full
set.seed(1)
# Membagi Data Menjadi Training dan Test ----
sampled_iris <- datairis %>%
mutate(training = sample(0:1,
nrow(datairis),
replace = TRUE))
training_data <- sampled_iris %>%
filter(training == 1)
test_data <- sampled_iris %>%
filter(training == 0)
# Model Regresi ----
model1 <-
datairis%>%
glm(SpeciesNum ~ SepalLengthCm + SepalWidthCm +
PetalLengthCm + PetalWidthCm ,data = .)
summary(model1)##
## Call:
## glm(formula = SpeciesNum ~ SepalLengthCm + SepalWidthCm + PetalLengthCm +
## PetalWidthCm, data = .)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.58569 -0.15004 0.02204 0.10421 0.50995
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.29668 0.22890 5.665 1.07e-07 ***
## SepalLengthCm -0.12292 0.06213 -1.979 0.05022 .
## SepalWidthCm -0.05584 0.06694 -0.834 0.40591
## PetalLengthCm 0.20505 0.06134 3.343 0.00112 **
## PetalWidthCm 0.67150 0.10125 6.632 1.08e-09 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 0.04732437)
##
## Null deviance: 80.795 on 121 degrees of freedom
## Residual deviance: 5.537 on 117 degrees of freedom
## AIC: -19.073
##
## Number of Fisher Scoring iterations: 2Confusion Matrix
classify <- function(probability){
ifelse( probability < 1.5, "setosa",
ifelse(probability < 2.5,
"versicolor",
"virginica"))
}
classified <- classify(predict(model1, datairis))
table(datairis$Species, classified, dnn= c("Data", "predictions"))## predictions
## Data setosa versicolor virginica
## Iris-setosa 38 0 0
## Iris-versicolor 0 39 2
## Iris-virginica 0 1 42confusion1 <-
table(datairis$Species, classified, dnn= c("Data", "predictions"))Akurasi Spesifikasi Sensitifiti
accuracy <- function(confusion_matrix) {
sum(diag(confusion_matrix))/sum(confusion_matrix)
}
specificity <- function(confusion_matrix) {
confusion_matrix[1,1]/
(confusion_matrix[1,1]+confusion_matrix[1,2])
}
sensitivity <- function(confusion_matrix) {
confusion_matrix[2,2]/
(confusion_matrix[2,1]+confusion_matrix[2,2])
}
prediction_summary <- function(confusion_matrix) {
c("accuracy" = accuracy(confusion_matrix),
"specificity" = specificity(confusion_matrix),
"sensitivity" = sensitivity(confusion_matrix))
}
prediction_summary(confusion1)## accuracy specificity sensitivity
## 0.9754098 1.0000000 1.0000000Model Dengan Sampling
set.seed(11)
# Model Regresi ----
model11 <-
training_data%>%
glm(SpeciesNum ~ SepalLengthCm + SepalWidthCm +
PetalLengthCm + PetalWidthCm ,data = .)
summary(model11)##
## Call:
## glm(formula = SpeciesNum ~ SepalLengthCm + SepalWidthCm + PetalLengthCm +
## PetalWidthCm, data = .)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.5726 -0.1626 0.0264 0.1129 0.4398
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.24361 0.33042 3.764 0.000393 ***
## SepalLengthCm -0.15243 0.09412 -1.620 0.110752
## SepalWidthCm -0.02441 0.10878 -0.224 0.823271
## PetalLengthCm 0.28509 0.09786 2.913 0.005073 **
## PetalWidthCm 0.52110 0.16591 3.141 0.002651 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 0.06094456)
##
## Null deviance: 38.9841 on 62 degrees of freedom
## Residual deviance: 3.5348 on 58 degrees of freedom
## AIC: 9.3159
##
## Number of Fisher Scoring iterations: 2classified11 <- classify(predict(model11, test_data))
table(test_data$Species, classified11, dnn= c("Data", "predictions"))## predictions
## Data setosa versicolor virginica
## Iris-setosa 22 0 0
## Iris-versicolor 0 18 0
## Iris-virginica 0 0 19confusion11 <-
table(test_data$Species, classified11, dnn= c("Data", "predictions"))
prediction_summary(confusion11)## accuracy specificity sensitivity
## 1 1 1Cross Validations
set.seed(111)
random_group <-
function(n, probs) {
probs <- probs / sum(probs)
g <- findInterval(seq(0, 1, length = n),
c(0, cumsum(probs)),
rightmost.closed = TRUE)
names(probs)[sample(g)]
}
partition <-
function(df, n, probs) {
replicate(n,
split(df, random_group(nrow(df),
probs)),
FALSE)
}
prediction_accuracy_iris <-
function(test_and_training) {
confusion <- vector(mode = "list",
length = length(test_and_training))
akurasi_iris <- vector(mode = "list",
length = length(test_and_training))
for (i in seq_along(test_and_training)) {
training <-
test_and_training[[i]]$training
test <- test_and_training[[i]]$test
model <- training %>%
glm(SpeciesNum ~ SepalLengthCm + SepalWidthCm +
PetalLengthCm + PetalWidthCm ,data = .)
classifications <-
classify(predict(model, test))
targets <- test$Species
confusion[[i]] <- table(test$Species, classifications, dnn= c("Data", "predictions"))
akurasi_iris[[i]] <- prediction_summary(confusion[[i]])
}
return(list(confusion, akurasi_iris))
}
datairis %>%
partition(5, c(training = 0.6, test = 0.4)) %>%
prediction_accuracy_iris## [[1]]
## [[1]][[1]]
## predictions
## Data setosa versicolor virginica
## Iris-setosa 11 0 0
## Iris-versicolor 0 15 1
## Iris-virginica 0 1 21
##
## [[1]][[2]]
## predictions
## Data setosa versicolor virginica
## Iris-setosa 17 0 0
## Iris-versicolor 0 16 0
## Iris-virginica 0 2 14
##
## [[1]][[3]]
## predictions
## Data setosa versicolor virginica
## Iris-setosa 16 0 0
## Iris-versicolor 0 16 0
## Iris-virginica 0 1 16
##
## [[1]][[4]]
## predictions
## Data setosa versicolor virginica
## Iris-setosa 15 0 0
## Iris-versicolor 0 17 2
## Iris-virginica 0 0 15
##
## [[1]][[5]]
## predictions
## Data setosa versicolor virginica
## Iris-setosa 12 0 0
## Iris-versicolor 0 20 1
## Iris-virginica 0 0 16
##
##
## [[2]]
## [[2]][[1]]
## accuracy specificity sensitivity
## 0.9591837 1.0000000 1.0000000
##
## [[2]][[2]]
## accuracy specificity sensitivity
## 0.9591837 1.0000000 1.0000000
##
## [[2]][[3]]
## accuracy specificity sensitivity
## 0.9795918 1.0000000 1.0000000
##
## [[2]][[4]]
## accuracy specificity sensitivity
## 0.9591837 1.0000000 1.0000000
##
## [[2]][[5]]
## accuracy specificity sensitivity
## 0.9795918 1.0000000 1.0000000regresi <-
datairis %>%
partition(5, c(training = 0.6, test = 0.4)) %>%
prediction_accuracy_iris
mean_accuracy <-
function(listdata) {
akurasi <- vector(mode = "list",
length = length(listdata))
for (i in seq_along(listdata)) {
akurasi[i] <- listdata[[2]][[i]][1]
}
akurasi <-
unlist(akurasi)
return(mean(akurasi))
}
mean_accuracy(regresi)## [1] 0.9693878regresi_accuracy <-
mean_accuracy(regresi)Decisions Tree
Memuat Package
library(party)Model Dataset Full
set.seed(2)
# Model Regresi Logistik ----
model2 <-
datairis%>%
ctree(SpeciesNum ~ SepalLengthCm + SepalLengthCm +
PetalLengthCm + PetalWidthCm, data = .)
summary(model2)## Length Class Mode
## 1 BinaryTree S4Confusion Matrix
classified2 <- classify(predict(model2, datairis))
table(datairis$Species, classified2, dnn= c("Data", "predictions"))## predictions
## Data setosa versicolor virginica
## Iris-setosa 38 0 0
## Iris-versicolor 0 40 1
## Iris-virginica 0 3 40confusion2 <-
table(datairis$Species, classified2, dnn= c("Data", "predictions"))Akurasi Spesifikasi Sensitifiti
prediction_summary(confusion2)## accuracy specificity sensitivity
## 0.9672131 1.0000000 1.0000000Model Dengan Sampling
set.seed(22)
# Model Regresi ----
model22 <-
training_data%>%
ctree(SpeciesNum ~ SepalLengthCm + SepalLengthCm+
PetalLengthCm + PetalWidthCm, data = .)
summary(model22)## Length Class Mode
## 1 BinaryTree S4classified22 <- classify(predict(model22, test_data))
table(test_data$Species, classified22, dnn= c("Data", "predictions"))## predictions
## Data setosa versicolor virginica
## Iris-setosa 21 1 0
## Iris-versicolor 0 17 1
## Iris-virginica 0 0 19confusion22 <-
table(test_data$Species, classified22, dnn= c("Data", "predictions"))
prediction_summary(confusion22)## accuracy specificity sensitivity
## 0.9661017 0.9545455 1.0000000Cross Validations
set.seed(222)
prediction_accuracy_iris <-
function(test_and_training) {
confusion <- vector(mode = "list",
length = length(test_and_training))
akurasi_iris <- vector(mode = "list",
length = length(test_and_training))
for (i in seq_along(test_and_training)) {
training <-
test_and_training[[i]]$training
test <- test_and_training[[i]]$test
model <- training %>%
ctree(SpeciesNum ~ SepalLengthCm + SepalLengthCm+
PetalLengthCm + PetalWidthCm, data = .)
classifications <-
classify(predict(model, test))
targets <- test$Species
confusion[[i]] <- table(test$Species, classifications, dnn= c("Data", "predictions"))
akurasi_iris[[i]] <- prediction_summary(confusion[[i]])
}
return(list(confusion, akurasi_iris))
}
datairis %>%
partition(5, c(training = 0.6, test = 0.4)) %>%
prediction_accuracy_iris## [[1]]
## [[1]][[1]]
## predictions
## Data setosa versicolor virginica
## Iris-setosa 12 0 0
## Iris-versicolor 0 13 1
## Iris-virginica 0 1 22
##
## [[1]][[2]]
## predictions
## Data setosa versicolor virginica
## Iris-setosa 15 1 0
## Iris-versicolor 0 15 1
## Iris-virginica 0 0 17
##
## [[1]][[3]]
## predictions
## Data setosa versicolor virginica
## Iris-setosa 14 0 0
## Iris-versicolor 0 16 2
## Iris-virginica 0 0 17
##
## [[1]][[4]]
## predictions
## Data setosa versicolor virginica
## Iris-setosa 11 2 0
## Iris-versicolor 0 20 0
## Iris-virginica 0 0 16
##
## [[1]][[5]]
## predictions
## Data setosa versicolor virginica
## Iris-setosa 15 0 0
## Iris-versicolor 0 18 1
## Iris-virginica 0 0 15
##
##
## [[2]]
## [[2]][[1]]
## accuracy specificity sensitivity
## 0.9591837 1.0000000 1.0000000
##
## [[2]][[2]]
## accuracy specificity sensitivity
## 0.9591837 0.9375000 1.0000000
##
## [[2]][[3]]
## accuracy specificity sensitivity
## 0.9591837 1.0000000 1.0000000
##
## [[2]][[4]]
## accuracy specificity sensitivity
## 0.9591837 0.8461538 1.0000000
##
## [[2]][[5]]
## accuracy specificity sensitivity
## 0.9795918 1.0000000 1.0000000Decision_tree <-
datairis %>%
partition(5, c(training = 0.6, test = 0.4)) %>%
prediction_accuracy_iris
mean_accuracy <-
function(listdata) {
akurasi <- vector(mode = "list",
length = length(listdata))
for (i in seq_along(listdata)) {
akurasi[i] <- listdata[[2]][[i]][1]
}
akurasi <-
unlist(akurasi)
return(mean(akurasi))
}
mean_accuracy(Decision_tree)## [1] 0.9387755tree_accuracy <-
mean_accuracy(Decision_tree)model2 %>%
plot
Random Forest
Memuat Package
library(randomForest)Model Dataset Full
set.seed(3)
# Model Regresi Logistik ----
model3 <-
datairis%>%
randomForest(SpeciesNum ~ SepalLengthCm + SepalLengthCm+
PetalLengthCm + PetalWidthCm, data = .)
summary(model3)## Length Class Mode
## call 3 -none- call
## type 1 -none- character
## predicted 122 -none- numeric
## mse 500 -none- numeric
## rsq 500 -none- numeric
## oob.times 122 -none- numeric
## importance 3 -none- numeric
## importanceSD 0 -none- NULL
## localImportance 0 -none- NULL
## proximity 0 -none- NULL
## ntree 1 -none- numeric
## mtry 1 -none- numeric
## forest 11 -none- list
## coefs 0 -none- NULL
## y 122 -none- numeric
## test 0 -none- NULL
## inbag 0 -none- NULL
## terms 3 terms callConfusion Matrix
classified3 <- classify(predict(model3, datairis))
table(datairis$Species, classified3, dnn= c("Data", "predictions"))## predictions
## Data setosa versicolor virginica
## Iris-setosa 38 0 0
## Iris-versicolor 0 40 1
## Iris-virginica 0 1 42confusion3 <-
table(datairis$Species, classified3, dnn= c("Data", "predictions"))Akurasi Spesifikasi Sensitifiti
prediction_summary(confusion3)## accuracy specificity sensitivity
## 0.9836066 1.0000000 1.0000000Model Dengan Sampling
set.seed(33)
# Model Regresi ----
model33 <-
training_data%>%
randomForest(SpeciesNum ~ SepalLengthCm + SepalLengthCm+
PetalLengthCm + PetalWidthCm, data = .)## Warning in randomForest.default(m, y, ...): The response has five or fewer
## unique values. Are you sure you want to do regression?summary(model33)## Length Class Mode
## call 3 -none- call
## type 1 -none- character
## predicted 63 -none- numeric
## mse 500 -none- numeric
## rsq 500 -none- numeric
## oob.times 63 -none- numeric
## importance 3 -none- numeric
## importanceSD 0 -none- NULL
## localImportance 0 -none- NULL
## proximity 0 -none- NULL
## ntree 1 -none- numeric
## mtry 1 -none- numeric
## forest 11 -none- list
## coefs 0 -none- NULL
## y 63 -none- numeric
## test 0 -none- NULL
## inbag 0 -none- NULL
## terms 3 terms callclassified33 <- classify(predict(model33, test_data))
table(test_data$Species, classified33, dnn= c("Data", "predictions"))## predictions
## Data setosa versicolor virginica
## Iris-setosa 22 0 0
## Iris-versicolor 0 17 1
## Iris-virginica 0 0 19confusion33 <-
table(test_data$Species, classified33, dnn= c("Data", "predictions"))
prediction_summary(confusion33)## accuracy specificity sensitivity
## 0.9830508 1.0000000 1.0000000Cross Validations
set.seed(333)
prediction_accuracy_iris <-
function(test_and_training) {
confusion <- vector(mode = "list",
length = length(test_and_training))
akurasi_iris <- vector(mode = "list",
length = length(test_and_training))
for (i in seq_along(test_and_training)) {
training <-
test_and_training[[i]]$training
test <- test_and_training[[i]]$test
model <- training %>%
randomForest(SpeciesNum ~ SepalLengthCm + SepalLengthCm+
PetalLengthCm + PetalWidthCm, data = .)
classifications <-
classify(predict(model, test))
targets <- test$Species
confusion[[i]] <- table(test$Species, classifications, dnn= c("Data", "predictions"))
akurasi_iris[[i]] <- prediction_summary(confusion[[i]])
}
return(list(confusion, akurasi_iris))
}
datairis %>%
partition(5, c(training = 0.6, test = 0.4)) %>%
prediction_accuracy_iris## [[1]]
## [[1]][[1]]
## predictions
## Data setosa versicolor virginica
## Iris-setosa 14 0 0
## Iris-versicolor 0 14 1
## Iris-virginica 0 3 17
##
## [[1]][[2]]
## predictions
## Data setosa versicolor virginica
## Iris-setosa 13 0 0
## Iris-versicolor 0 15 1
## Iris-virginica 0 1 19
##
## [[1]][[3]]
## predictions
## Data setosa versicolor virginica
## Iris-setosa 15 0 0
## Iris-versicolor 0 17 3
## Iris-virginica 0 0 14
##
## [[1]][[4]]
## predictions
## Data setosa versicolor virginica
## Iris-setosa 12 0 0
## Iris-versicolor 0 17 1
## Iris-virginica 0 1 18
##
## [[1]][[5]]
## predictions
## Data setosa versicolor virginica
## Iris-setosa 20 0 0
## Iris-versicolor 0 14 2
## Iris-virginica 0 1 12
##
##
## [[2]]
## [[2]][[1]]
## accuracy specificity sensitivity
## 0.9183673 1.0000000 1.0000000
##
## [[2]][[2]]
## accuracy specificity sensitivity
## 0.9591837 1.0000000 1.0000000
##
## [[2]][[3]]
## accuracy specificity sensitivity
## 0.9387755 1.0000000 1.0000000
##
## [[2]][[4]]
## accuracy specificity sensitivity
## 0.9591837 1.0000000 1.0000000
##
## [[2]][[5]]
## accuracy specificity sensitivity
## 0.9387755 1.0000000 1.0000000Random_Forest <-
datairis %>%
partition(5, c(training = 0.6, test = 0.4)) %>%
prediction_accuracy_iris
mean_accuracy <-
function(listdata) {
akurasi <- vector(mode = "list",
length = length(listdata))
for (i in seq_along(listdata)) {
akurasi[i] <- listdata[[2]][[i]][1]
}
akurasi <-
unlist(akurasi)
return(mean(akurasi))
}
mean_accuracy(Random_Forest)## [1] 0.9591837Random_Forest_accuracy <-
mean_accuracy(Random_Forest)Tabel Akurasi dari Ketiga Metode
tribble(~ Regresi, ~ Tree, ~ Random_Forest,
regresi_accuracy, tree_accuracy, Random_Forest_accuracy)## # A tibble: 1 x 3
## Regresi Tree Random_Forest
## <dbl> <dbl> <dbl>
## 1 0.969 0.939 0.959Akurasi Tertinggi adalah Regresi.
Mencoba Menebak Jenis Bunga Berdasarkan Data Submission
Dataset
submission## Id SepalLengthCm SepalWidthCm PetalLengthCm PetalWidthCm
## 1 136 4.6 3.6 1.0 0.2
## 2 137 5.0 3.2 1.2 0.2
## 3 138 5.8 4.0 1.2 0.2
## 4 139 5.3 3.7 1.5 0.2
## 5 140 5.4 3.7 1.5 0.2
## 6 141 5.6 2.9 3.6 1.3
## 7 142 5.5 2.3 4.0 1.3
## 8 143 5.2 2.7 3.9 1.4
## 9 144 6.6 3.0 4.4 1.4
## 10 145 6.7 3.1 4.4 1.4
## 11 146 7.2 3.2 6.0 1.8
## 12 147 7.3 2.9 6.3 1.8
## 13 148 6.3 2.5 5.0 1.9
## 14 149 6.3 2.8 5.1 1.5
## 15 150 6.2 2.8 4.8 1.8Hasil Classifications
classify(predict(model1, submission))## 1 2 3 4 5 6
## "setosa" "setosa" "setosa" "setosa" "setosa" "versicolor"
## 7 8 9 10 11 12
## "versicolor" "versicolor" "versicolor" "versicolor" "virginica" "virginica"
## 13 14 15
## "virginica" "versicolor" "virginica"submission %>%
mutate( classify(predict(model1, submission)))## Id SepalLengthCm SepalWidthCm PetalLengthCm PetalWidthCm
## 1 136 4.6 3.6 1.0 0.2
## 2 137 5.0 3.2 1.2 0.2
## 3 138 5.8 4.0 1.2 0.2
## 4 139 5.3 3.7 1.5 0.2
## 5 140 5.4 3.7 1.5 0.2
## 6 141 5.6 2.9 3.6 1.3
## 7 142 5.5 2.3 4.0 1.3
## 8 143 5.2 2.7 3.9 1.4
## 9 144 6.6 3.0 4.4 1.4
## 10 145 6.7 3.1 4.4 1.4
## 11 146 7.2 3.2 6.0 1.8
## 12 147 7.3 2.9 6.3 1.8
## 13 148 6.3 2.5 5.0 1.9
## 14 149 6.3 2.8 5.1 1.5
## 15 150 6.2 2.8 4.8 1.8
## classify(predict(model1, submission))
## 1 setosa
## 2 setosa
## 3 setosa
## 4 setosa
## 5 setosa
## 6 versicolor
## 7 versicolor
## 8 versicolor
## 9 versicolor
## 10 versicolor
## 11 virginica
## 12 virginica
## 13 virginica
## 14 versicolor
## 15 virginica