S's PlayGround

Logo

Testing & Educational Purpose

import pandas as pd

data = pd.read_csv(r"C:\Users\Lena\Documents\PASCA\Datasets\Kaggle - Fatalities in Palestine Israel\archive\fatalities_isr_pse_conflict_2000_to_2023.csv")

data.head()
name date_of_event age citizenship event_location event_location_district event_location_region date_of_death gender took_part_in_the_hostilities place_of_residence place_of_residence_district type_of_injury ammunition killed_by notes
0 'Abd a-Rahman Suleiman Muhammad Abu Daghash 2023-09-24 32.0 Palestinian Nur Shams R.C. Tulkarm West Bank 2023-09-24 M NaN Nur Shams R.C. Tulkarm gunfire live ammunition Israeli security forces Fatally shot by Israeli forces while standing ...
1 Usayed Farhan Muhammad 'Ali Abu 'Ali 2023-09-24 21.0 Palestinian Nur Shams R.C. Tulkarm West Bank 2023-09-24 M NaN Nur Shams R.C. Tulkarm gunfire live ammunition Israeli security forces Fatally shot by Israeli forces while trying to...
2 'Abdallah 'Imad Sa'ed Abu Hassan 2023-09-22 16.0 Palestinian Kfar Dan Jenin West Bank 2023-09-22 M NaN al-Yamun Jenin gunfire live ammunition Israeli security forces Fatally shot by soldiers while firing at them ...
3 Durgham Muhammad Yihya al-Akhras 2023-09-20 19.0 Palestinian 'Aqbat Jaber R.C. Jericho West Bank 2023-09-20 M NaN 'Aqbat Jaber R.C. Jericho gunfire live ammunition Israeli security forces Shot in the head by Israeli forces while throw...
4 Raafat 'Omar Ahmad Khamaisah 2023-09-19 15.0 Palestinian Jenin R.C. Jenin West Bank 2023-09-19 M NaN Jenin Jenin gunfire live ammunition Israeli security forces Wounded by soldiers’ gunfire after running awa... 
missing_values = data[['age', 'type_of_injury', 'killed_by', 'citizenship']].isnull().sum()
missing_values

age               129
type_of_injury    291
killed_by           0
citizenship         0
dtype: int64

median_age = data['age'].median()
data['age'].fillna(median_age, inplace=True)

mode_type_of_injury = data['type_of_injury'].mode()[0]
data['type_of_injury'].fillna(mode_type_of_injury, inplace=True)

updated_missing_values = data[['age', 'type_of_injury', 'killed_by', 'citizenship']].isnull().sum()
updated_missing_values

age               0
type_of_injury    0
killed_by         0
citizenship       0
dtype: int64

ammunition_distribution = data['ammunition'].value_counts()
type_of_injury_distribution = data['type_of_injury'].value_counts()
killed_by_distribution = data['killed_by'].value_counts()

ammunition_distribution, type_of_injury_distribution, killed_by_distribution

(missile                        2877
 live ammunition                1514
 shell                           675
 explosive belt                  326
 bomb                            249
 mortar fire                      51
 knife                            37
 flechette shells                 22
 rubber-coated metal bullets      19
 0.22-caliber bullets             16
 phosphorus shell                 16
 Qassam rocket                    15
 car bomb                         15
 teargas canister                 13
 rocket                           12
 grad rocket                       7
 sponge rounds                     2
 grenade                           2
 flare bomb                        1
 stun grenade                      1
 rock                              1
 Name: ammunition, dtype: int64,
 gunfire                         10140
 explosion                         555
 shelling                          311
 stabbing                           48
 house demolition                   25
 hit by a vehicle                   18
 beating                             9
 stones throwing                     6
 being bludgeoned with an axe        4
 fire                                4
 physically assaulted                2
 physical assault                    1
 Strangulation                       1
 Name: type_of_injury, dtype: int64,
 Israeli security forces    10000
 Palestinian civilians       1028
 Israeli civilians             96
 Name: killed_by, dtype: int64)

#Pembuatan model prediksi
from sklearn.preprocessing import OneHotEncoder

features = data[['ammunition', 'type_of_injury']]
target = data['killed_by']

encoder = OneHotEncoder(sparse=False)
features_encoded = encoder.fit_transform(features)

encoded_feature_names = encoder.get_feature_names_out(['ammunition', 'type_of_injury'])
features_encoded_df = pd.DataFrame(features_encoded, columns=encoded_feature_names)

features_encoded_df.head()
ammunition_0.22-caliber bullets ammunition_Qassam rocket ammunition_bomb ammunition_car bomb ammunition_explosive belt ammunition_flare bomb ammunition_flechette shells ammunition_grad rocket ammunition_grenade ammunition_knife ... type_of_injury_explosion type_of_injury_fire type_of_injury_gunfire type_of_injury_hit by a vehicle type_of_injury_house demolition type_of_injury_physical assault type_of_injury_physically assaulted type_of_injury_shelling type_of_injury_stabbing type_of_injury_stones throwing
0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 ... 0.0 0.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 ... 0.0 0.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 ... 0.0 0.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 ... 0.0 0.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 ... 0.0 0.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

5 rows × 35 columns

#Implementasi regresi logistik
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LogisticRegression
from sklearn.metrics import classification_report

X_train, X_test, y_train, y_test = train_test_split(features_encoded_df, target, test_size=0.3, random_state=42)

log_reg_model = LogisticRegression(max_iter=1000)
log_reg_model.fit(X_train, y_train)

y_pred = log_reg_model.predict(X_test)

classification_report_results = classification_report(y_test, y_pred)
classification_report_results

C:\Users\Lena\anaconda3\lib\site-packages\sklearn\metrics\_classification.py:1318: 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))
C:\Users\Lena\anaconda3\lib\site-packages\sklearn\metrics\_classification.py:1318: 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))
C:\Users\Lena\anaconda3\lib\site-packages\sklearn\metrics\_classification.py:1318: 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))





'                         precision    recall  f1-score   support\n\n      Israeli civilians       0.00      0.00      0.00        31\nIsraeli security forces       0.95      1.00      0.97      2986\n  Palestinian civilians       0.93      0.61      0.74       321\n\n               accuracy                           0.95      3338\n              macro avg       0.63      0.54      0.57      3338\n           weighted avg       0.94      0.95      0.94      3338\n'

#Diagram ROC LR
from sklearn.metrics import roc_curve, auc, RocCurveDisplay
import matplotlib.pyplot as plt

fpr = dict()
tpr = dict()
roc_auc = dict()
n_classes = y_train.nunique()
classes = log_reg_model.classes_

y_test_bin = pd.get_dummies(y_test, columns=classes).to_numpy()

y_score = log_reg_model.predict_proba(X_test)

for i in range(n_classes):
    fpr[i], tpr[i], _ = roc_curve(y_test_bin[:, i], y_score[:, i])
    roc_auc[i] = auc(fpr[i], tpr[i])

plt.figure(figsize=(10, 8))

colors = ['blue', 'red', 'green']
for i, color in zip(range(n_classes), colors):
    plt.plot(fpr[i], tpr[i], color=color, lw=2,
             label='ROC curve of class {0} (area = {1:0.2f})'
             ''.format(classes[i], roc_auc[i]))

plt.plot([0, 1], [0, 1], 'k--', lw=2)
plt.xlim([0.0, 1.0])
plt.ylim([0.0, 1.05])
plt.xlabel('False Positive Rate')
plt.ylabel('True Positive Rate')
plt.title('Receiver Operating Characteristic (ROC) for Logistic Regression')
plt.legend(loc="lower right")
plt.show()

png

#Confusion matrix LR
from sklearn.metrics import confusion_matrix
import seaborn as sns

conf_mat = confusion_matrix(y_test, y_pred)

plt.figure(figsize=(10, 8))
sns.heatmap(conf_mat, annot=True, fmt="d", 
            xticklabels=log_reg_model.classes_,
            yticklabels=log_reg_model.classes_)
plt.title('Confusion Matrix for Logistic Regression')
plt.ylabel('True label')
plt.xlabel('Predicted label')
plt.show()

png

#Implementasi SVM
from sklearn.svm import SVC

svm_model = SVC()
svm_model.fit(X_train, y_train)

y_pred_svm = svm_model.predict(X_test)

classification_report_svm = classification_report(y_test, y_pred_svm)
classification_report_svm

C:\Users\Lena\anaconda3\lib\site-packages\sklearn\metrics\_classification.py:1318: 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))
C:\Users\Lena\anaconda3\lib\site-packages\sklearn\metrics\_classification.py:1318: 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))
C:\Users\Lena\anaconda3\lib\site-packages\sklearn\metrics\_classification.py:1318: 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))





'                         precision    recall  f1-score   support\n\n      Israeli civilians       0.00      0.00      0.00        31\nIsraeli security forces       0.95      0.99      0.97      2986\n  Palestinian civilians       0.92      0.63      0.75       321\n\n               accuracy                           0.95      3338\n              macro avg       0.62      0.54      0.57      3338\n           weighted avg       0.94      0.95      0.94      3338\n'

#Diagram ROC SVM
from sklearn.preprocessing import label_binarize

y_test_bin_svm = label_binarize(y_test, classes=classes)

y_score_svm = svm_model.decision_function(X_test)

fpr_svm = dict()
tpr_svm = dict()
roc_auc_svm = dict()

for i in range(n_classes):
    fpr_svm[i], tpr_svm[i], _ = roc_curve(y_test_bin_svm[:, i], y_score_svm[:, i])
    roc_auc_svm[i] = auc(fpr_svm[i], tpr_svm[i])

plt.figure(figsize=(10, 8))

for i, color in zip(range(n_classes), colors):
    plt.plot(fpr_svm[i], tpr_svm[i], color=color, lw=2,
             label='ROC curve of class {0} (area = {1:0.2f})'
             ''.format(classes[i], roc_auc_svm[i]))

plt.plot([0, 1], [0, 1], 'k--', lw=2)
plt.xlim([0.0, 1.0])
plt.ylim([0.0, 1.05])
plt.xlabel('False Positive Rate')
plt.ylabel('True Positive Rate')
plt.title('Receiver Operating Characteristic (ROC) for SVM')
plt.legend(loc="lower right")
plt.show()

png

#Confusion Matrix SVM
conf_mat_svm = confusion_matrix(y_test, y_pred_svm)

plt.figure(figsize=(10, 8))
sns.heatmap(conf_mat_svm, annot=True, fmt="d", 
            xticklabels=svm_model.classes_,
            yticklabels=svm_model.classes_)
plt.title('Confusion Matrix for SVM')
plt.ylabel('True label')
plt.xlabel('Predicted label')
plt.show()

png

#Perbandingan LR dan SVM
from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score

accuracy_lr = accuracy_score(y_test, y_pred)
precision_lr = precision_score(y_test, y_pred, average='weighted')
recall_lr = recall_score(y_test, y_pred, average='weighted')
f1_score_lr = f1_score(y_test, y_pred, average='weighted')

accuracy_svm = accuracy_score(y_test, y_pred_svm)
precision_svm = precision_score(y_test, y_pred_svm, average='weighted')
recall_svm = recall_score(y_test, y_pred_svm, average='weighted')
f1_score_svm = f1_score(y_test, y_pred_svm, average='weighted')

performance_comparison = pd.DataFrame({
    'Metric': ['Accuracy', 'Precision', 'Recall', 'F1 Score'],
    'Logistic Regression': [accuracy_lr, precision_lr, recall_lr, f1_score_lr],
    'SVM': [accuracy_svm, precision_svm, recall_svm, f1_score_svm]
})

performance_comparison.set_index('Metric', inplace=True)
performance_comparison


C:\Users\Lena\anaconda3\lib\site-packages\sklearn\metrics\_classification.py:1318: UndefinedMetricWarning: Precision is 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))
C:\Users\Lena\anaconda3\lib\site-packages\sklearn\metrics\_classification.py:1318: UndefinedMetricWarning: Precision is 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 SVM
Metric
Accuracy 0.949970 0.950270
Precision 0.940563 0.940493
Recall 0.949970 0.950270
F1 Score 0.941623 0.942244 
#Rubah menjadi persentase
performance_comparison_percentage = performance_comparison * 100
performance_comparison_percentage
Logistic Regression SVM
Metric
Accuracy 94.997004 95.026962
Precision 94.056332 94.049348
Recall 94.997004 95.026962
F1 Score 94.162253 94.224373 
#Distribusi amunisi
plt.figure(figsize=(12, 8))
sns.countplot(y=data['ammunition'], order = data['ammunition'].value_counts().index)
plt.title('Distribusi Data Jenis Amunisi')
plt.xlabel('Jumlah')
plt.ylabel('Jenis Amunisi')
plt.show()

png

#Distribusi killed by
plt.figure(figsize=(10, 6))
sns.countplot(y=data['killed_by'], order=data['killed_by'].value_counts().index)
plt.title('Distribusi Data "Dibunuh Oleh"')
plt.xlabel('Jumlah')
plt.ylabel('Dibunuh Oleh')
plt.show()

png

#Distribusi jenis cedera
plt.figure(figsize=(12, 8))
sns.countplot(y=data['type_of_injury'], order=data['type_of_injury'].value_counts().index)
plt.title('Distribusi Data Jenis Cedera')
plt.xlabel('Jumlah')
plt.ylabel('Jenis Cedera')
plt.show()

png