Muhammad Sandhi Khadafi/2KB04/(20122007)

Import Dataset
# Import library
import pandas as pd
import numpy as np

# import dataset
data = pd.read_csv('https://gitlab.com/andreass.bayu/file-directory/-/raw/
main/adult.csv')

# Lihat 5 data awal

data.head(5)

age workclass fnlwgt education educational-num marital-status

\
0 25 Private 226802 11th 7 Never-married

1 38 Private 89814 HS-grad 9 Married-civ-spouse

2 28 Local-gov 336951 Assoc-acdm 12 Married-civ-spouse

3 44 Private 160323 Some-college 10 Married-civ-spouse

4 18 ? 103497 Some-college 10 Never-married

occupation relationship race gender capital-gain capital-loss

\
0 Machine-op-inspct Own-child Black Male 0 0

1 Farming-fishing Husband White Male 0 0

2 Protective-serv Husband White Male 0 0

3 Machine-op-inspct Husband Black Male 7688 0

4 ? Own-child White Female 0 0

hours-per-week native-country income

0 40 United-States <=50K
1 50 United-States <=50K
2 40 United-States >50K
3 40 United-States >50K
4 30 United-States <=50K

Review Dataset
# Lihat deskripsi dari tiap kolom
data.describe()

age fnlwgt educational-num capital-gain \

count 48842.000000 4.884200e+04 48842.000000 48842.000000
mean 38.643585 1.896641e+05 10.078089 1079.067626
std 13.710510 1.056040e+05 2.570973 7452.019058
min 17.000000 1.228500e+04 1.000000 0.000000
25% 28.000000 1.175505e+05 9.000000 0.000000
50% 37.000000 1.781445e+05 10.000000 0.000000
75% 48.000000 2.376420e+05 12.000000 0.000000
max 90.000000 1.490400e+06 16.000000 99999.000000

capital-loss hours-per-week
count 48842.000000 48842.000000
mean 87.502314 40.422382
std 403.004552 12.391444
min 0.000000 1.000000
25% 0.000000 40.000000
50% 0.000000 40.000000
75% 0.000000 45.000000
max 4356.000000 99.000000

# Lihat tipe data dari tiap kolom

data.dtypes

age int64
workclass object
fnlwgt int64
education object
educational-num int64
marital-status object
occupation object
relationship object
race object
gender object
capital-gain int64
capital-loss int64
hours-per-week int64
native-country object
income object
dtype: object

# mlihat jumlah atribut dan data / dimensi data

data.shape
(48842, 15)

# hitung dan melihat jumlah data per label kelas

for col in data.columns:
if data[col].dtype == "object":
print('Attribute name:',col)
print('-------------------')
print(data[col].value_counts())
print('-------------------')

Attribute name: workclass

-------------------
Private 33906
Self-emp-not-inc 3862
Local-gov 3136
? 2799
State-gov 1981
Self-emp-inc 1695
Federal-gov 1432
Without-pay 21
Never-worked 10
Name: workclass, dtype: int64
-------------------
Attribute name: education
-------------------
HS-grad 15784
Some-college 10878
Bachelors 8025
Masters 2657
Assoc-voc 2061
11th 1812
Assoc-acdm 1601
10th 1389
7th-8th 955
Prof-school 834
9th 756
12th 657
Doctorate 594
5th-6th 509
1st-4th 247
Preschool 83
Name: education, dtype: int64
-------------------
Attribute name: marital-status
-------------------
Married-civ-spouse 22379
Never-married 16117
Divorced 6633
Separated 1530
Widowed 1518
Married-spouse-absent 628
Married-AF-spouse 37
Name: marital-status, dtype: int64
-------------------
Attribute name: occupation
-------------------
Prof-specialty 6172
Craft-repair 6112
Exec-managerial 6086
Adm-clerical 5611
Sales 5504
Other-service 4923
Machine-op-inspct 3022
? 2809
Transport-moving 2355
Handlers-cleaners 2072
Farming-fishing 1490
Tech-support 1446
Protective-serv 983
Priv-house-serv 242
Armed-Forces 15
Name: occupation, dtype: int64
-------------------
Attribute name: relationship
-------------------
Husband 19716
Not-in-family 12583
Own-child 7581
Unmarried 5125
Wife 2331
Other-relative 1506
Name: relationship, dtype: int64
-------------------
Attribute name: race
-------------------
White 41762
Black 4685
Asian-Pac-Islander 1519
Amer-Indian-Eskimo 470
Other 406
Name: race, dtype: int64
-------------------
Attribute name: gender
-------------------
Male 32650
Female 16192
Name: gender, dtype: int64
-------------------
Attribute name: native-country
-------------------
United-States 43832
Mexico 951
? 857
Philippines 295
Germany 206
Puerto-Rico 184
Canada 182
El-Salvador 155
India 151
Cuba 138
England 127
China 122
South 115
Jamaica 106
Italy 105
Dominican-Republic 103
Japan 92
Guatemala 88
Poland 87
Vietnam 86
Columbia 85
Haiti 75
Portugal 67
Taiwan 65
Iran 59
Greece 49
Nicaragua 49
Peru 46
Ecuador 45
France 38
Ireland 37
Hong 30
Thailand 30
Cambodia 28
Trinadad&Tobago 27
Laos 23
Yugoslavia 23
Outlying-US(Guam-USVI-etc) 23
Scotland 21
Honduras 20
Hungary 19
Holand-Netherlands 1
Name: native-country, dtype: int64
-------------------
Attribute name: income
-------------------
<=50K 37155
>50K 11687
Name: income, dtype: int64
-------------------
#import library seaborn untuk visualisasi
import seaborn as sns
import matplotlib.pyplot as plt
%matplotlib inline

# Plot figure untuk menentukan distribusi kelas

plt.figure(figsize=(8,5))

# menghitung baris setiap kelas

sns.countplot(x="income", data=data)

<matplotlib.axes._subplots.AxesSubplot at 0x7fa6995b1f90>

Dataset memiliki distribusi kelas yang tidak seimbang/imbalanced sehingga secara teknis akan
digunakan teknik untuk menangani data yang tidak seimbang.

Dataset preparation
# Buat salinan dataframe
df = data.copy(deep = True)

# mengubah/convert nilai "?" nilai ke bentuk Na / NaN untuk diproses lebih

lanjut
for col in data.columns:
df[[col]] = data[[col]].replace('?',np.NaN)

# seleksi kolom fitur/feature columns dari dataset

null_data = df.iloc[:,:-1]
# temukan nilai null untuk semua atribut dan jumlahkan total nilai null
null_data.isnull().sum()

age 0
workclass 2799
fnlwgt 0
education 0
educational-num 0
marital-status 0
occupation 2809
relationship 0
race 0
gender 0
capital-gain 0
capital-loss 0
hours-per-week 0
native-country 857
dtype: int64

# jatuhkan/drop semua baris yang memiliki nilai null

df = df.dropna()

# pilih kolom fitur/feature columns dari dataset

null_data = df.iloc[:,:-1]

# cek ulang nilai null

null_data.isnull().sum()

age 0
workclass 0
fnlwgt 0
education 0
educational-num 0
marital-status 0
occupation 0
relationship 0
race 0
gender 0
capital-gain 0
capital-loss 0
hours-per-week 0
native-country 0
dtype: int64

StandardScaler adalah class dari sklearn untuk melakukan normalisasi data agar data yang
digunakan tidak memiliki penyimpangan yang besar.
# Import library standard scaler

from sklearn.preprocessing import StandardScaler

# Buat dataframe dengan tipe data int64

colname= []
for col in df.columns:
if df[col].dtype == "int64":
colname.append(col)

# Buat salinan dataset untuk keperluan persiapan data / data preparation

df_copy = df.copy(deep = True)
df_fe = df.copy()

# Buat kerangka data untuk fitur kategoris / categorical features

df_fe.drop('income',axis='columns', inplace=True)
df_fe.drop(colname,axis='columns', inplace=True)

# buat dataframe untuk kelas target / target class

df_cl = df.copy()
df_cl.drop(df_copy.iloc[:,:-1],axis='columns', inplace=True)

# membuat objek scaler / scaler object

std_scaler = StandardScaler()
std_scaler

# Normalisasikan atribut numerik dan tetapkan ke dalam dataframe baru

df_norm = pd.DataFrame(std_scaler.fit_transform(df_copy[colname]),
columns=colname)

# import library Ordinal Encoder dari package library sklearn.preprocessing

from sklearn.preprocessing import OrdinalEncoder
ord_enc = OrdinalEncoder()

# enconde fitur kategoris/categorical features menjadi fitur

numerik/numerical features
for col in df_fe.columns[:]:
if df_fe[col].dtype == "object":
df_fe[col] = ord_enc.fit_transform(df_fe[[col]])

# enconde label kategorikal/categorical label menjadi label biner/binary

label
df_cl["income"] = np.where(df_cl["income"].str.contains(">50K"), 0, 1)

# Masukkan kolom id ke datasets yang berbeda

df_norm.insert(0, 'id', range(0, 0 + len(df_norm)))
df_fe.insert(0, 'id', range(0, 0 + len(df_fe)))
df_cl.insert(0, 'id', range(0, 0 + len(df_cl)))

# Lihat shapes datasets yang telah di proses

print(df_norm.shape)
print(df_fe.shape)
print(df_cl.shape)

(45222, 7)
(45222, 9)
(45222, 2)

# Gabungkan semua datasets

df_feature = pd.merge(df_norm,df_fe, on=["id"])
df_final = pd.merge(df_feature,df_cl, on=["id"])

# Drop kolom id dari gabungan dataset

df_final.drop('id',axis='columns', inplace=True)

# Lihat 5 data awal dari gabungan dataset

df_final.head(5)

age fnlwgt educational-num capital-gain capital-loss \

0 -1.024983 0.350889 -1.221559 -0.146733 -0.21878
1 -0.041455 -0.945878 -0.438122 -0.146733 -0.21878
2 -0.798015 1.393592 0.737034 -0.146733 -0.21878
3 0.412481 -0.278420 -0.046403 0.877467 -0.21878
4 -0.344079 0.084802 -1.613277 -0.146733 -0.21878

hours-per-week workclass education marital-status occupation \

0 -0.078120 2.0 1.0 4.0 6.0
1 0.754701 2.0 11.0 2.0 4.0
2 -0.078120 1.0 7.0 2.0 10.0
3 -0.078120 2.0 15.0 2.0 6.0
4 -0.910942 2.0 0.0 4.0 7.0

relationship race gender native-country income

0 3.0 2.0 1.0 38.0 1
1 0.0 4.0 1.0 38.0 1
2 0.0 4.0 1.0 38.0 0
3 0.0 2.0 1.0 38.0 0
4 1.0 4.0 1.0 38.0 1
Visualization
p = df_final.hist(figsize = (20,20))

Scatter matrix plot adalah plot yang digunakan untuk membuat sekumpulan scatter plot
dari beberapa pasang variabel. Hal ini sangat bermanfaat terutama ketika ingin
menganalisis bagaimana bentuk hubungan antar variabel. Plot ini sangat bermanfaat untuk
digunakan untuk data yang ukurannya tidak terlalu besar. Untuk menggunakan scatter
matrix kita harus memanggil fungsi scatter_matrix dari pandas.plotting
from pandas.plotting import scatter_matrix

p=scatter_matrix(df_final,figsize=(25, 25))
# Buat visualisasi korelasi data dengan heatmap
import seaborn as sns
import matplotlib.pyplot as plt

# plot heatmap
plt.figure(figsize=(12,10))
p=sns.heatmap(df_final.corr(), annot=True,cmap ='RdYlGn')
Proses Modelling dengan KNN
import numpy as np
from sklearn.model_selection import KFold

X=df_final.iloc[:,:].to_numpy()
y=df_final.iloc[:,-1:].to_numpy()
kf = KFold(n_splits=5)
kf.get_n_splits(X)

print(kf)

for train_index, test_index in kf.split(X):

print("TRAIN:", train_index, "TEST:", test_index)
x_train, x_test = X[train_index], X[test_index]
y_train, y_test = y[train_index], y[test_index]
KFold(n_splits=5, random_state=None, shuffle=False)
TRAIN: [ 9045 9046 9047 ... 45219 45220 45221] TEST: [ 0 1 2 ...
9042 9043 9044]
TRAIN: [ 0 1 2 ... 45219 45220 45221] TEST: [ 9045 9046 9047 ...
18087 18088 18089]
TRAIN: [ 0 1 2 ... 45219 45220 45221] TEST: [18090 18091 18092 ...
27131 27132 27133]
TRAIN: [ 0 1 2 ... 45219 45220 45221] TEST: [27134 27135 27136 ...
36175 36176 36177]
TRAIN: [ 0 1 2 ... 36175 36176 36177] TEST: [36178 36179 36180 ...
45219 45220 45221]

print('-------- x axis test ----------')

print(x_test)
print('-------- x axis train ---------')
print(x_train)
print('-------- y axis test ----------')
print(y_test)
print('-------- y axis train ---------')
print(y_train)
print('*******************************')

-------- x axis test ----------

[[ 0.63944887 1.45787715 -0.43812161 ... 1. 38.
1. ]
[ 1.24469679 -1.49349376 -0.046403 ... 1. 38.
1. ]
[-0.26842301 -0.14603391 -0.046403 ... 1. 38.
1. ]
...
[ 1.47166476 -0.35805983 -0.43812161 ... 0. 38.
1. ]
[-1.25195088 0.11127873 -0.43812161 ... 1. 38.
1. ]
[ 1.01772882 0.92951628 -0.43812161 ... 0. 38.
0. ]]
-------- x axis train ---------
[[-1.02498291 0.35088942 -1.22155881 ... 1. 38.
1. ]
[-0.04145504 -0.94587846 -0.43812161 ... 1. 38.
1. ]
[-0.79801494 1.39359159 0.73703421 ... 1. 38.
0. ]
...
[-0.344079 -0.77568407 -0.43812161 ... 1. 10.
0. ]
[ 0.86641684 0.04524191 1.52047141 ... 1. 37.
1. ]
[-0.64670296 0.84029651 -0.43812161 ... 0. 38.
1. ]]
-------- y axis test ----------
[[1]
[1]
[1]
...
[1]
[1]
[0]]
-------- y axis train ---------
[[1]
[1]
[0]
...
[0]
[1]
[1]]
*******************************

# import library model KNN dengan alias/as 'KNeighborsClassifier'

from sklearn.neighbors import KNeighborsClassifier
import sklearn.metrics as metrics
import matplotlib.pyplot as plt

# buat variabel kosong untuk menyimpan metrik KNN/KNN metrics

scores=[]
# Kita coba nilai k yang berbeda untuk KNN (dari k=1 sampai k=26)
lrange=list(range(1,20))
# loop proses KNN
for k in lrange:
# masukkan nilai k dan ukuran 'jarak'
knn=KNeighborsClassifier(n_neighbors=k)
# masukan data train/ data latih untuk melatih KNN
knn.fit(x_train,y_train.ravel())
# lihat prediksi KNN dengan memasukkan data uji/data test
y_pred=knn.predict(x_test)
# tambahkan performance metric akurasi
scores.append(metrics.accuracy_score(y_test,y_pred))
plt.figure(2,figsize=(15,5))

optimal_k = lrange[scores.index(max(scores))]
print("Nilai k KNN yang optimal adalah %d" % optimal_k)
print("Skor optimalnya adalah %.2f" % max(scores))

# plot hasilnya
plt.plot(lrange, scores,ls='dashed')
plt.xlabel('Nilai dari k untuk KNN')
plt.ylabel('Accuracy Score')
plt.title('Accuracy Scores untuk nilai k dari k-Nearest-Neighbors')
plt.show()
Nilai k KNN yang optimal adalah 3
Skor optimalnya adalah 0.95

#EVALUASI HASIL MATRIKS

y_pred = knn.predict(x_test)
from sklearn import metrics
cnf_matrix = metrics.confusion_matrix(y_test, y_pred)
p = sns.heatmap(pd.DataFrame(cnf_matrix), annot=True, cmap='PuRd' ,fmt='g')
plt.title('Confusion matrix', y=1.1)
plt.ylabel('Actual label')
plt.xlabel('Predicted label')

Text(0.5, 15.0, 'Predicted label')

#import classification_report
from sklearn.metrics import classification_report
print(classification_report(y_test,y_pred))

precision recall f1-score support

0 0.91 0.79 0.85 2303

1 0.93 0.97 0.95 6741

accuracy 0.93 9044

macro avg 0.92 0.88 0.90 9044
weighted avg 0.93 0.93 0.92 9044