#Import libraries
import matplotlib.pyplot as plt
import warnings
import pandas as pd
import seaborn as sns
import numpy as np
import random
from sklearn.preprocessing import StandardScaler
from sklearn.preprocessing import MinMaxScaler
#naive bayes
from sklearn.naive_bayes import GaussianNB
from sklearn import metrics
#knn
from sklearn.neighbors import KNeighborsClassifier
#decision tree
from sklearn.tree import DecisionTreeClassifier
#svm
from sklearn import svm
#logistic regression
from sklearn.linear_model import LogisticRegression
from google.colab import drive
drive.mount("/content/drive")
Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount("/content/drive", force_remount=True).
cd drive/
/content/drive
%cd MyDrive/Colab\ Notebooks/
/content/drive/MyDrive/Colab Notebooks
cd tubesML2/
[Errno 2] No such file or directory: 'tubesML2/' /content/drive/MyDrive/Colab Notebooks
!ls
kendaraan_test.csv kendaraan_train.csv malin visdat
DT = 'kendaraan_test.csv'
datatest = pd.read_csv(DT)
datatest.head()
| Jenis_Kelamin | Umur | SIM | Kode_Daerah | Sudah_Asuransi | Umur_Kendaraan | Kendaraan_Rusak | Premi | Kanal_Penjualan | Lama_Berlangganan | Tertarik | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | Wanita | 49 | 1 | 8 | 0 | 1-2 Tahun | Pernah | 46963 | 26 | 145 | 0 |
| 1 | Pria | 22 | 1 | 47 | 1 | < 1 Tahun | Tidak | 39624 | 152 | 241 | 0 |
| 2 | Pria | 24 | 1 | 28 | 1 | < 1 Tahun | Tidak | 110479 | 152 | 62 | 0 |
| 3 | Pria | 46 | 1 | 8 | 1 | 1-2 Tahun | Tidak | 36266 | 124 | 34 | 0 |
| 4 | Pria | 35 | 1 | 23 | 0 | 1-2 Tahun | Pernah | 26963 | 152 | 229 | 0 |
datatest.dtypes
Jenis_Kelamin object Umur int64 SIM int64 Kode_Daerah int64 Sudah_Asuransi int64 Umur_Kendaraan object Kendaraan_Rusak object Premi int64 Kanal_Penjualan int64 Lama_Berlangganan int64 Tertarik int64 dtype: object
datatest.shape
(47639, 11)
DT = 'kendaraan_train.csv'
datatrain = pd.read_csv(DT)
datatrain.head()
| id | Jenis_Kelamin | Umur | SIM | Kode_Daerah | Sudah_Asuransi | Umur_Kendaraan | Kendaraan_Rusak | Premi | Kanal_Penjualan | Lama_Berlangganan | Tertarik | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 1 | Wanita | 30.0 | 1.0 | 33.0 | 1.0 | < 1 Tahun | Tidak | 28029.0 | 152.0 | 97.0 | 0 |
| 1 | 2 | Pria | 48.0 | 1.0 | 39.0 | 0.0 | > 2 Tahun | Pernah | 25800.0 | 29.0 | 158.0 | 0 |
| 2 | 3 | NaN | 21.0 | 1.0 | 46.0 | 1.0 | < 1 Tahun | Tidak | 32733.0 | 160.0 | 119.0 | 0 |
| 3 | 4 | Wanita | 58.0 | 1.0 | 48.0 | 0.0 | 1-2 Tahun | Tidak | 2630.0 | 124.0 | 63.0 | 0 |
| 4 | 5 | Pria | 50.0 | 1.0 | 35.0 | 0.0 | > 2 Tahun | NaN | 34857.0 | 88.0 | 194.0 | 0 |
datatrain.dtypes
id int64 Jenis_Kelamin object Umur float64 SIM float64 Kode_Daerah float64 Sudah_Asuransi float64 Umur_Kendaraan object Kendaraan_Rusak object Premi float64 Kanal_Penjualan float64 Lama_Berlangganan float64 Tertarik int64 dtype: object
datatrain.shape
(285831, 12)
datatest.isnull().sum()
Jenis_Kelamin 0 Umur 0 SIM 0 Kode_Daerah 0 Sudah_Asuransi 0 Umur_Kendaraan 0 Kendaraan_Rusak 0 Premi 0 Kanal_Penjualan 0 Lama_Berlangganan 0 Tertarik 0 dtype: int64
datatest = datatest.dropna(how='any')
datatest.isnull().sum()
Jenis_Kelamin 0 Umur 0 SIM 0 Kode_Daerah 0 Sudah_Asuransi 0 Umur_Kendaraan 0 Kendaraan_Rusak 0 Premi 0 Kanal_Penjualan 0 Lama_Berlangganan 0 Tertarik 0 dtype: int64
data sudah tidak memiliki missing value.
datatest.describe
<bound method NDFrame.describe of Jenis_Kelamin Umur SIM ... Kanal_Penjualan Lama_Berlangganan Tertarik 0 Wanita 49 1 ... 26 145 0 1 Pria 22 1 ... 152 241 0 2 Pria 24 1 ... 152 62 0 3 Pria 46 1 ... 124 34 0 4 Pria 35 1 ... 152 229 0 ... ... ... ... ... ... ... ... 47634 Pria 61 1 ... 124 67 0 47635 Pria 41 1 ... 157 232 0 47636 Pria 24 1 ... 152 211 0 47637 Pria 59 1 ... 26 239 1 47638 Pria 52 1 ... 124 170 0 [47639 rows x 11 columns]>
datatrain.isnull().sum()
id 0 Jenis_Kelamin 14440 Umur 14214 SIM 14404 Kode_Daerah 14306 Sudah_Asuransi 14229 Umur_Kendaraan 14275 Kendaraan_Rusak 14188 Premi 14569 Kanal_Penjualan 14299 Lama_Berlangganan 13992 Tertarik 0 dtype: int64
datatrain = datatrain.dropna(how='any')
datatrain.isnull().sum()
id 0 Jenis_Kelamin 0 Umur 0 SIM 0 Kode_Daerah 0 Sudah_Asuransi 0 Umur_Kendaraan 0 Kendaraan_Rusak 0 Premi 0 Kanal_Penjualan 0 Lama_Berlangganan 0 Tertarik 0 dtype: int64
data sudah tidak memiliki missing value.
datatrain.describe
<bound method NDFrame.describe of id Jenis_Kelamin ... Lama_Berlangganan Tertarik 0 1 Wanita ... 97.0 0 1 2 Pria ... 158.0 0 3 4 Wanita ... 63.0 0 5 6 Pria ... 171.0 0 8 9 Wanita ... 31.0 0 ... ... ... ... ... ... 285826 285827 Wanita ... 217.0 0 285827 285828 Wanita ... 50.0 0 285828 285829 Wanita ... 226.0 0 285829 285830 Pria ... 270.0 0 285830 285831 Pria ... 44.0 0 [171068 rows x 12 columns]>
datatest.replace(['< 1 Tahun','1-2 Tahun', '> 2 Tahun'],[1,2,3],inplace=True)
datatest.replace(['Pria', 'Wanita'],[1,0],inplace=True)
datatest.replace(['Pernah', 'Tidak'],[1,0],inplace=True)
datatest.head()
| Jenis_Kelamin | Umur | SIM | Kode_Daerah | Sudah_Asuransi | Umur_Kendaraan | Kendaraan_Rusak | Premi | Kanal_Penjualan | Lama_Berlangganan | Tertarik | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 0 | 49 | 1 | 8 | 0 | 2 | 1 | 46963 | 26 | 145 | 0 |
| 1 | 1 | 22 | 1 | 47 | 1 | 1 | 0 | 39624 | 152 | 241 | 0 |
| 2 | 1 | 24 | 1 | 28 | 1 | 1 | 0 | 110479 | 152 | 62 | 0 |
| 3 | 1 | 46 | 1 | 8 | 1 | 2 | 0 | 36266 | 124 | 34 | 0 |
| 4 | 1 | 35 | 1 | 23 | 0 | 2 | 1 | 26963 | 152 | 229 | 0 |
#split data test untuk label
testtarget = datatest.iloc[:,[10]]
testData = datatest.iloc[:,[0,1,2,3,4,5,6,7,8,9]]
testData.head()
| Jenis_Kelamin | Umur | SIM | Kode_Daerah | Sudah_Asuransi | Umur_Kendaraan | Kendaraan_Rusak | Premi | Kanal_Penjualan | Lama_Berlangganan | |
|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 0 | 49 | 1 | 8 | 0 | 2 | 1 | 46963 | 26 | 145 |
| 1 | 1 | 22 | 1 | 47 | 1 | 1 | 0 | 39624 | 152 | 241 |
| 2 | 1 | 24 | 1 | 28 | 1 | 1 | 0 | 110479 | 152 | 62 |
| 3 | 1 | 46 | 1 | 8 | 1 | 2 | 0 | 36266 | 124 | 34 |
| 4 | 1 | 35 | 1 | 23 | 0 | 2 | 1 | 26963 | 152 | 229 |
#normalisasi data dengan minmaxscaler
Ndatatest = MinMaxScaler(feature_range=(0,1)).fit_transform(testData)
Ndatatest
array([[0. , 0.44615385, 1. , ..., 0.08247463, 0.15432099,
0.46712803],
[1. , 0.03076923, 1. , ..., 0.06882157, 0.93209877,
0.79930796],
[1. , 0.06153846, 1. , ..., 0.20063624, 0.93209877,
0.1799308 ],
...,
[1. , 0.06153846, 1. , ..., 0.05668654, 0.93209877,
0.69550173],
[1. , 0.6 , 1. , ..., 0.06540597, 0.15432099,
0.79238754],
[1. , 0.49230769, 1. , ..., 0. , 0.75925926,
0.55363322]])
datatrain.replace(['< 1 Tahun','1-2 Tahun', '> 2 Tahun'],[1,2,3],inplace=True)
datatrain.replace(['Pria', 'Wanita'],[1,0],inplace=True)
datatrain.replace(['Pernah', 'Tidak'],[1,0],inplace=True)
datatrain.head()
| id | Jenis_Kelamin | Umur | SIM | Kode_Daerah | Sudah_Asuransi | Umur_Kendaraan | Kendaraan_Rusak | Premi | Kanal_Penjualan | Lama_Berlangganan | Tertarik | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 1 | 0 | 30.0 | 1.0 | 33.0 | 1.0 | 1 | 0 | 28029.0 | 152.0 | 97.0 | 0 |
| 1 | 2 | 1 | 48.0 | 1.0 | 39.0 | 0.0 | 3 | 1 | 25800.0 | 29.0 | 158.0 | 0 |
| 3 | 4 | 0 | 58.0 | 1.0 | 48.0 | 0.0 | 2 | 0 | 2630.0 | 124.0 | 63.0 | 0 |
| 5 | 6 | 1 | 21.0 | 1.0 | 35.0 | 1.0 | 1 | 0 | 22735.0 | 152.0 | 171.0 | 0 |
| 8 | 9 | 0 | 20.0 | 1.0 | 8.0 | 1.0 | 1 | 0 | 30786.0 | 160.0 | 31.0 | 0 |
#split data test untuk label
traintarget = datatrain.iloc[:,[11]]
trainData = datatrain.iloc[:,[1,2,3,4,5,6,7,8,9,10]]
trainData.head()
| Jenis_Kelamin | Umur | SIM | Kode_Daerah | Sudah_Asuransi | Umur_Kendaraan | Kendaraan_Rusak | Premi | Kanal_Penjualan | Lama_Berlangganan | |
|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 0 | 30.0 | 1.0 | 33.0 | 1.0 | 1 | 0 | 28029.0 | 152.0 | 97.0 |
| 1 | 1 | 48.0 | 1.0 | 39.0 | 0.0 | 3 | 1 | 25800.0 | 29.0 | 158.0 |
| 3 | 0 | 58.0 | 1.0 | 48.0 | 0.0 | 2 | 0 | 2630.0 | 124.0 | 63.0 |
| 5 | 1 | 21.0 | 1.0 | 35.0 | 1.0 | 1 | 0 | 22735.0 | 152.0 | 171.0 |
| 8 | 0 | 20.0 | 1.0 | 8.0 | 1.0 | 1 | 0 | 30786.0 | 160.0 | 31.0 |
#normalisasi data dengan minmaxscaler
Ndatatrain = MinMaxScaler(feature_range=(0,1)).fit_transform(trainData)
Ndatatrain
array([[0. , 0.15384615, 1. , ..., 0.04725088, 0.93209877,
0.30103806],
[1. , 0.43076923, 1. , ..., 0.04310417, 0.17283951,
0.51211073],
[0. , 0.58461538, 1. , ..., 0. , 0.75925926,
0.183391 ],
...,
[0. , 0.04615385, 1. , ..., 0.08766127, 0.93209877,
0.74740484],
[1. , 0.73846154, 1. , ..., 0.05185337, 0.75925926,
0.89965398],
[1. , 0.38461538, 1. , ..., 0.06297264, 0.15432099,
0.11764706]])
accuracyArr = []
confusionmatrixArr = []
classificationreportArr = []
# Naive Bayes
#train model dan predict
NaiveBayes = GaussianNB().fit(Ndatatrain,traintarget)
NaiveBayes
/usr/local/lib/python3.7/dist-packages/sklearn/naive_bayes.py:206: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel(). y = column_or_1d(y, warn=True)
GaussianNB(priors=None, var_smoothing=1e-09)
# Naive Bayes
#menggunakan model untuk memprediksi test set
#returns array dari predicted results
NaiveBayesPredict = NaiveBayes.predict(Ndatatest)
NaiveBayesPredict
array([1, 0, 0, ..., 0, 1, 0])
# Naive Bayes
confusionmatrix = metrics.confusion_matrix(testtarget, NaiveBayesPredict)
classificationreport = metrics.classification_report(testtarget, NaiveBayesPredict)
accuracy = metrics.accuracy_score(testtarget, NaiveBayesPredict)
confusionmatrixArr.append(confusionmatrix)
classificationreportArr.append(classificationreport)
accuracyArr.append(accuracy)
print("Confusion Matrix:","\n",confusionmatrix)
print("\nClassification Report:\n",classificationreport)
print("Accuracy:", accuracy)
Confusion Matrix:
[[24824 16954]
[ 148 5713]]
Classification Report:
precision recall f1-score support
0 0.99 0.59 0.74 41778
1 0.25 0.97 0.40 5861
accuracy 0.64 47639
macro avg 0.62 0.78 0.57 47639
weighted avg 0.90 0.64 0.70 47639
Accuracy: 0.6410084174730788
# KNN
#train model, disini value knya yaitu 19
k = 19
#train model dan predict
KNN = KNeighborsClassifier(n_neighbors = k).fit(Ndatatrain,traintarget)
KNN
/usr/local/lib/python3.7/dist-packages/ipykernel_launcher.py:6: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().
KNeighborsClassifier(algorithm='auto', leaf_size=30, metric='minkowski',
metric_params=None, n_jobs=None, n_neighbors=19, p=2,
weights='uniform')
# KNN
#menggunakan model untuk memprediksi test set
KNNPredict = KNN.predict(Ndatatest)
KNNPredict
array([0, 0, 0, ..., 0, 0, 0])
# KNN
confusionmatrix = metrics.confusion_matrix(testtarget, KNNPredict)
classificationreport = metrics.classification_report(testtarget, KNNPredict)
accuracy = metrics.accuracy_score(testtarget, KNNPredict)
confusionmatrixArr.append(confusionmatrix)
classificationreportArr.append(classificationreport)
accuracyArr.append(accuracy)
print("Confusion Matrix:","\n",confusionmatrix)
print("\nClassification Report:\n",classificationreport)
print("\nAccuracy:", accuracy)
Confusion Matrix:
[[41152 626]
[ 5477 384]]
Classification Report:
precision recall f1-score support
0 0.88 0.99 0.93 41778
1 0.38 0.07 0.11 5861
accuracy 0.87 47639
macro avg 0.63 0.53 0.52 47639
weighted avg 0.82 0.87 0.83 47639
Accuracy: 0.8718906778059993
# Decision Tree
#train model dan predict
DesicionTree = DecisionTreeClassifier().fit(Ndatatrain,traintarget)
DesicionTree
DecisionTreeClassifier(ccp_alpha=0.0, class_weight=None, criterion='gini',
max_depth=None, max_features=None, max_leaf_nodes=None,
min_impurity_decrease=0.0, min_impurity_split=None,
min_samples_leaf=1, min_samples_split=2,
min_weight_fraction_leaf=0.0, presort='deprecated',
random_state=None, splitter='best')
# Decision Tree
#menggunakan model untuk memprediksi test set
DesicionTreePredict = DesicionTree.predict(Ndatatest)
DesicionTreePredict
array([0, 0, 0, ..., 0, 0, 0])
# Decision Tree
confusionmatrix = metrics.confusion_matrix(testtarget, DesicionTreePredict)
classificationreport = metrics.classification_report(testtarget, DesicionTreePredict)
accuracy = metrics.accuracy_score(testtarget, DesicionTreePredict)
confusionmatrixArr.append(confusionmatrix)
classificationreportArr.append(classificationreport)
accuracyArr.append(accuracy)
print("Confusion Matrix:","\n",confusionmatrix)
print("\nClassification Report:\n",classificationreport)
print("\nAccuracy:", accuracy)
Confusion Matrix:
[[37377 4401]
[ 4049 1812]]
Classification Report:
precision recall f1-score support
0 0.90 0.89 0.90 41778
1 0.29 0.31 0.30 5861
accuracy 0.82 47639
macro avg 0.60 0.60 0.60 47639
weighted avg 0.83 0.82 0.82 47639
Accuracy: 0.8226243204097483
# Support Vector Machine
#train model dan predict
SupportMachineVector = svm.SVC(kernel='linear').fit(Ndatatrain,traintarget)
SupportMachineVector
/usr/local/lib/python3.7/dist-packages/sklearn/utils/validation.py:760: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel(). y = column_or_1d(y, warn=True)
SVC(C=1.0, break_ties=False, cache_size=200, class_weight=None, coef0=0.0,
decision_function_shape='ovr', degree=3, gamma='scale', kernel='linear',
max_iter=-1, probability=False, random_state=None, shrinking=True,
tol=0.001, verbose=False)
# Support Vector Machine
#menggunakan model untuk memprediksi test set
SupportMachineVectorPredict = SupportMachineVector.predict(Ndatatest)
SupportMachineVectorPredict
array([0, 0, 0, ..., 0, 0, 0])
# Support Vector Machine
confusionmatrix = metrics.confusion_matrix(testtarget, SupportMachineVectorPredict)
classificationreport = metrics.classification_report(testtarget, SupportMachineVectorPredict)
accuracy = metrics.accuracy_score(testtarget, SupportMachineVectorPredict)
confusionmatrixArr.append(confusionmatrix)
classificationreportArr.append(classificationreport)
accuracyArr.append(accuracy)
print("Confusion Matrix:","\n",confusionmatrix)
print("\nClassification Report:\n",classificationreport)
print("\nAccuracy:", accuracy)
Confusion Matrix:
[[41778 0]
[ 5861 0]]
Classification Report:
precision recall f1-score support
0 0.88 1.00 0.93 41778
1 0.00 0.00 0.00 5861
accuracy 0.88 47639
macro avg 0.44 0.50 0.47 47639
weighted avg 0.77 0.88 0.82 47639
Accuracy: 0.8769705493398267
/usr/local/lib/python3.7/dist-packages/sklearn/metrics/_classification.py:1272: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior. _warn_prf(average, modifier, msg_start, len(result))
# Logistic Regression
#train model dan predict
LogiscticRegression = LogisticRegression().fit(Ndatatrain,traintarget)
LogiscticRegression
/usr/local/lib/python3.7/dist-packages/sklearn/utils/validation.py:760: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel(). y = column_or_1d(y, warn=True)
LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=True,
intercept_scaling=1, l1_ratio=None, max_iter=100,
multi_class='auto', n_jobs=None, penalty='l2',
random_state=None, solver='lbfgs', tol=0.0001, verbose=0,
warm_start=False)
# Logistic Regression
#menggunakan model untuk memprediksi test set
LogiscticRegressionPredict = LogiscticRegression.predict(Ndatatest)
LogiscticRegressionPredict
array([0, 0, 0, ..., 0, 0, 0])
# Logistic Regression
confusionmatrix = metrics.confusion_matrix(testtarget, LogiscticRegressionPredict)
classificationreport = metrics.classification_report(testtarget, LogiscticRegressionPredict)
accuracy = metrics.accuracy_score(testtarget, LogiscticRegressionPredict)
confusionmatrixArr.append(confusionmatrix)
classificationreportArr.append(classificationreport)
accuracyArr.append(accuracy)
print("Confusion Matrix:","\n",confusionmatrix)
print("\nClassification Report:\n",classificationreport)
print("\nAccuracy:", accuracy)
Confusion Matrix:
[[41738 40]
[ 5840 21]]
Classification Report:
precision recall f1-score support
0 0.88 1.00 0.93 41778
1 0.34 0.00 0.01 5861
accuracy 0.88 47639
macro avg 0.61 0.50 0.47 47639
weighted avg 0.81 0.88 0.82 47639
Accuracy: 0.8765717164508071
ArrMethod = ['Naive Bayes','KNN','Desicion Tree','SVM','Logistic Regression']
accuracyV = pd.DataFrame(accuracyArr, columns=['Akurasi'])
accuracyV['Metode']=ArrMethod
accuracyV = accuracyV.sort_values(by='Akurasi', ascending=False)
accuracyV
| Akurasi | Metode | |
|---|---|---|
| 3 | 0.876971 | SVM |
| 4 | 0.876572 | Logistic Regression |
| 1 | 0.871891 | KNN |
| 2 | 0.822624 | Desicion Tree |
| 0 | 0.641008 | Naive Bayes |
# Akurasi
warnings.filterwarnings('ignore')
plt.rcParams['figure.figsize'] = (10, 10)
plt.style.use('seaborn')
sns.set(rc={'figure.facecolor':'white'})
sns.set_palette('Accent')
fig, (ax1) = plt.subplots(1)
fig.suptitle('Comparison', fontsize=20, fontweight='bold')
fig.set_figheight(10)
fig.set_figwidth(20)
fig.set_facecolor('white')
sns.barplot(data=accuracyV, x='Metode',y='Akurasi',palette = "Reds")
<matplotlib.axes._subplots.AxesSubplot at 0x7f8aee6bcad0>
