Hugging Face's logo

Spaces:
SerifeGokdas
/
Fake-Real-New-Classification
Sleeping

App Files Community
SerifeGokdas commited on
Commit
dfb3177
·
verified ·
1 Parent(s): 3010ef4

created project

Browse files
Files changed (11) hide show
  1. app.py +136 -0
  2. logistic_regression.ipynb +141 -0
  3. lstm_model.ipynb +116 -0
  4. neural_networks.ipynb +139 -0
  5. pre_processing.ipynb +363 -0
  6. random_forest.ipynb +141 -0
  7. requirements.txt +14 -0
  8. support_vector_machine.ipynb +139 -0
  9. tokenized_processing.ipynb +222 -0
  10. vector_processing.ipynb +87 -0
  11. xg_boost.ipynb +149 -0
app.py ADDED
@@ -0,0 +1,136 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ import streamlit as st
2
+ import pandas as pd
3
+ import numpy as np
4
+ from sklearn.preprocessing import LabelEncoder
5
+ import joblib
6
+ import tensorflow as tf
7
+ from sentence_transformers import SentenceTransformer
8
+ import torch
9
+ import re
10
+ from nltk.corpus import stopwords
11
+ from nltk.stem import WordNetLemmatizer
12
+ import nltk
13
+ nltk.download('stopwords')
14
+ nltk.download('wordnet')
15
+ from transformers import AutoTokenizer
16
+
17
+
18
+ # Tokenizer'ı başlatıyoruz
19
+ tokenizer = AutoTokenizer.from_pretrained("alibayram/tr_tokenizer", use_fast=True)
20
+ embedding_model = SentenceTransformer('sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2')
21
+
22
+ # Model dosyaları
23
+ log_reg_model = joblib.load('logistic_regression_model.pkl') # Logistic Regression modeli
24
+ #lstm_model = tf.keras.models.load_model('lstm_model.h5') # LSTM model
25
+ nn_model = joblib.load('neural_network_model.pkl') # Neural network modeli
26
+ rf_model = joblib.load('random_forest_model.pkl') # Random Forest modeli
27
+ svm_model = joblib.load('support_vector_machine_model.pkl') # SVM modeli
28
+ xgboost_model = joblib.load('xgboost_model.pkl') # XGBoost modeli
29
+
30
+
31
+
32
+ # Preprocessing ve veri hazırlığı için fonksiyonlar
33
+ def preprocess_text(text):
34
+ STOPWORDS = set(stopwords.words('turkish'))
35
+ STOPWORDS.add('mi')
36
+ # Küçük harfe çevirme
37
+ text = text.lower()
38
+ # Noktalama işaretlerini kaldır
39
+ text = re.sub(r'[^\w\s]', '', text)
40
+ # Stopwords kaldır ve lemmatization uygula
41
+ text = " ".join([word for word in text.split() if word not in STOPWORDS])
42
+ return text
43
+
44
+ def tokenize_text(text):
45
+
46
+ tokens = tokenizer.tokenize(text)
47
+ return tokens
48
+
49
+ def vectorize_text(text):
50
+ # Tokenized listeyi string'e dönüştürelim
51
+ titles_as_strings = [' '.join(eval(tokens)) for tokens in text]
52
+ # Metni vektöre dönüştürme
53
+ return embedding_model.encode(titles_as_strings)
54
+
55
+ def vectorize_text(text):
56
+
57
+ # Her bir başlığı string olarak kabul ederek işleme alıyoruz
58
+ titles_as_strings = [' '.join(tokens) if isinstance(tokens, list) else tokens for tokens in text]
59
+
60
+ # Metni vektöre dönüştürme
61
+ return embedding_model.encode(titles_as_strings)
62
+
63
+
64
+
65
+ # Tahmin fonksiyonları
66
+ def predict_logistic_regression(text):
67
+
68
+ return log_reg_model.predict(text)
69
+
70
+
71
+
72
+ def predict_neural_network(text):
73
+
74
+ return nn_model.predict(text)
75
+
76
+ def predict_random_forest(text):
77
+
78
+ return rf_model.predict(text)
79
+
80
+ def predict_svm(text):
81
+
82
+ return svm_model.predict(text)
83
+
84
+ def predict_xgboost(text):
85
+
86
+ return xgboost_model.predict(text)
87
+
88
+ # Streamlit arayüzü
89
+ st.title("Haber Başlığı Doğruluk Tahmin Aracı")
90
+
91
+ # Kullanıcıdan haber başlığını alma
92
+ news_title = st.text_input("Haber Başlığını Girin:")
93
+
94
+ title_preprocesed=preprocess_text(news_title)
95
+ title_tokenized=tokenize_text(title_preprocesed)
96
+ title_vectorized=vectorize_text(title_tokenized)
97
+
98
+ # Model Seçim kutusu
99
+ model_choice = st.selectbox(
100
+ "Hangi modeli kullanmak istersiniz?",
101
+ ("Logistic Regression", "LSTM","Neural Network", "Random Forest", "Support Vector Machine", "XGBoost")
102
+ )
103
+
104
+ # "Tahmin Et" butonu
105
+ if st.button("Tahmin Et"):
106
+ if news_title:
107
+ # Seçilen modele göre tahmin yapma
108
+ if model_choice == "Logistic Regression":
109
+ prediction = predict_logistic_regression(title_vectorized)
110
+ elif model_choice == "Neural Network":
111
+ prediction = predict_neural_network(title_vectorized)
112
+ elif model_choice == "Random Forest":
113
+ prediction = predict_random_forest(title_vectorized)
114
+ elif model_choice == "Support Vector Machine":
115
+ prediction = predict_svm(title_vectorized)
116
+ elif model_choice == "XGBoost":
117
+ prediction = predict_xgboost(title_vectorized)
118
+
119
+
120
+
121
+ # Tahmin Sonucunu Gösterme
122
+ if prediction[0] == 1:
123
+ st.write(f"**Tahmin:** Bu haber başlığı doğru.")
124
+ else:
125
+ st.write(f"**Tahmin:** Bu haber başlığı yanlış.")
126
+
127
+ # Tüm modellerin tahminlerini yazma
128
+ st.write("### Tüm Modellerin Tahminleri:")
129
+ st.write(f"**Logistic Regression Tahmini:** {'Doğru' if predict_logistic_regression(title_vectorized)[0] == 1 else 'Yanlış'}")
130
+ st.write(f"**Neural Network Tahmini:** {'Doğru' if predict_neural_network(title_vectorized)[0] == 1 else 'Yanlış'}")
131
+ st.write(f"**Random Forest Tahmini:** {'Doğru' if predict_random_forest(title_vectorized)[0] == 1 else 'Yanlış'}")
132
+ st.write(f"**SVM Tahmini:** {'Doğru' if predict_svm(title_vectorized)[0] == 1 else 'Yanlış'}")
133
+ st.write(f"**XGBoost Tahmini:** {'Doğru' if predict_xgboost(title_vectorized)[0] == 1 else 'Yanlış'}")
134
+
135
+ else:
136
+ st.write("Lütfen bir haber başlığı girin.")
logistic_regression.ipynb ADDED
@@ -0,0 +1,141 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cells": [
3
+ {
4
+ "cell_type": "code",
5
+ "execution_count": 1,
6
+ "metadata": {},
7
+ "outputs": [
8
+ {
9
+ "name": "stdout",
10
+ "output_type": "stream",
11
+ "text": [
12
+ "Logistic Regression Accuracy: 0.88\n",
13
+ "\n",
14
+ "Classification Report:\n",
15
+ " precision recall f1-score support\n",
16
+ "\n",
17
+ " 0 0.90 0.85 0.88 1301\n",
18
+ " 1 0.85 0.90 0.88 1244\n",
19
+ "\n",
20
+ " accuracy 0.88 2545\n",
21
+ " macro avg 0.88 0.88 0.88 2545\n",
22
+ "weighted avg 0.88 0.88 0.88 2545\n",
23
+ "\n"
24
+ ]
25
+ },
26
+ {
27
+ "data": {
28
+ "image/png": "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",
29
+ "text/plain": [
30
+ "<Figure size 800x600 with 2 Axes>"
31
+ ]
32
+ },
33
+ "metadata": {},
34
+ "output_type": "display_data"
35
+ },
36
+ {
37
+ "data": {
38
+ "image/png": "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",
39
+ "text/plain": [
40
+ "<Figure size 800x600 with 1 Axes>"
41
+ ]
42
+ },
43
+ "metadata": {},
44
+ "output_type": "display_data"
45
+ },
46
+ {
47
+ "data": {
48
+ "text/plain": [
49
+ "['logistic_regression_model.pkl']"
50
+ ]
51
+ },
52
+ "execution_count": 1,
53
+ "metadata": {},
54
+ "output_type": "execute_result"
55
+ }
56
+ ],
57
+ "source": [
58
+ "import pandas as pd\n",
59
+ "from sklearn.model_selection import train_test_split\n",
60
+ "from sklearn.linear_model import LogisticRegression\n",
61
+ "from sklearn.metrics import accuracy_score, classification_report, confusion_matrix, roc_auc_score, roc_curve\n",
62
+ "import seaborn as sns\n",
63
+ "import matplotlib.pyplot as plt\n",
64
+ "import joblib\n",
65
+ "\n",
66
+ "# Veriyi yükleme\n",
67
+ "data = pd.read_csv('data_vectorized.csv')\n",
68
+ "\n",
69
+ "# Veriyi ayırma (özellikler ve etiketler)\n",
70
+ "X = data.drop(columns=['Label']) # Etiket hariç tüm sütunlar\n",
71
+ "y = data['Label'] # Etiket\n",
72
+ "\n",
73
+ "# Veriyi eğitim ve test olarak ayırma\n",
74
+ "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n",
75
+ "\n",
76
+ "# Logistic Regression modeli oluşturma ve eğitim\n",
77
+ "log_reg = LogisticRegression(max_iter=1000)\n",
78
+ "log_reg.fit(X_train, y_train)\n",
79
+ "\n",
80
+ "# Test seti üzerinde tahmin yapma\n",
81
+ "y_pred = log_reg.predict(X_test)\n",
82
+ "\n",
83
+ "# Doğruluk hesaplama\n",
84
+ "accuracy = accuracy_score(y_test, y_pred)\n",
85
+ "print(f\"Logistic Regression Accuracy: {accuracy:.2f}\")\n",
86
+ "\n",
87
+ "# Precision, Recall, F1-Score\n",
88
+ "print(\"\\nClassification Report:\")\n",
89
+ "print(classification_report(y_test, y_pred))\n",
90
+ "\n",
91
+ "# Confusion Matrix\n",
92
+ "conf_matrix = confusion_matrix(y_test, y_pred)\n",
93
+ "plt.figure(figsize=(8, 6))\n",
94
+ "sns.heatmap(conf_matrix, annot=True, fmt='d', cmap='Blues', xticklabels=[\"Class 0\", \"Class 1\"], yticklabels=[\"Class 0\", \"Class 1\"])\n",
95
+ "plt.title(\"Logistic Regression - Confusion Matrix\")\n",
96
+ "plt.xlabel(\"Predicted\")\n",
97
+ "plt.ylabel(\"Actual\")\n",
98
+ "plt.show()\n",
99
+ "\n",
100
+ "# ROC-AUC Hesaplama ve Grafik\n",
101
+ "if len(set(y_test)) == 2: # İkili sınıflandırma kontrolü\n",
102
+ " roc_auc = roc_auc_score(y_test, log_reg.predict_proba(X_test)[:, 1]) # Tahmin olasılıkları\n",
103
+ " fpr, tpr, thresholds = roc_curve(y_test, log_reg.predict_proba(X_test)[:, 1])\n",
104
+ " plt.figure(figsize=(8, 6))\n",
105
+ " plt.plot(fpr, tpr, label=f\"ROC-AUC: {roc_auc:.2f}\")\n",
106
+ " plt.plot([0, 1], [0, 1], 'k--', label=\"Random Guess\")\n",
107
+ " plt.xlabel(\"False Positive Rate\")\n",
108
+ " plt.ylabel(\"True Positive Rate\")\n",
109
+ " plt.title(\"Logistic Regression - ROC Curve\")\n",
110
+ " plt.legend()\n",
111
+ " plt.show()\n",
112
+ "\n",
113
+ "# Logistic Regression modelini kaydetme\n",
114
+ "joblib.dump(log_reg, 'logistic_regression_model.pkl')\n",
115
+ "\n",
116
+ "\n"
117
+ ]
118
+ }
119
+ ],
120
+ "metadata": {
121
+ "kernelspec": {
122
+ "display_name": "Python 3",
123
+ "language": "python",
124
+ "name": "python3"
125
+ },
126
+ "language_info": {
127
+ "codemirror_mode": {
128
+ "name": "ipython",
129
+ "version": 3
130
+ },
131
+ "file_extension": ".py",
132
+ "mimetype": "text/x-python",
133
+ "name": "python",
134
+ "nbconvert_exporter": "python",
135
+ "pygments_lexer": "ipython3",
136
+ "version": "3.10.10"
137
+ }
138
+ },
139
+ "nbformat": 4,
140
+ "nbformat_minor": 2
141
+ }
lstm_model.ipynb ADDED
@@ -0,0 +1,116 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cells": [
3
+ {
4
+ "cell_type": "code",
5
+ "execution_count": null,
6
+ "metadata": {
7
+ "vscode": {
8
+ "languageId": "plaintext"
9
+ }
10
+ },
11
+ "outputs": [],
12
+ "source": [
13
+ "import numpy as np\n",
14
+ "import pandas as pd\n",
15
+ "from sklearn.model_selection import train_test_split\n",
16
+ "from tensorflow.keras.models import Sequential\n",
17
+ "from tensorflow.keras.layers import Embedding, LSTM, Dense, Dropout\n",
18
+ "from tensorflow.keras.optimizers import Adam\n",
19
+ "from sklearn.preprocessing import LabelEncoder\n",
20
+ "from sklearn.metrics import accuracy_score, confusion_matrix, classification_report, roc_auc_score, roc_curve\n",
21
+ "import matplotlib.pyplot as plt\n",
22
+ "import joblib\n",
23
+ "\n",
24
+ "# Veriyi yükle\n",
25
+ "df = pd.read_csv('data_vectorized.csv')\n",
26
+ "\n",
27
+ "# Özellikler (X) ve Etiketler (y) ayırma\n",
28
+ "X = df.drop(columns=['Label']) # Özellikler (vectorized data)\n",
29
+ "y = df['Label'] # Etiketler (True/False)\n",
30
+ "\n",
31
+ "# Etiketleri sayılara dönüştürme (Label Encoding)\n",
32
+ "le = LabelEncoder()\n",
33
+ "y = le.fit_transform(y)\n",
34
+ "\n",
35
+ "# Eğitim ve test verilerini ayırma (%80 eğitim, %20 test)\n",
36
+ "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n",
37
+ "\n",
38
+ "# Model Parametreleri\n",
39
+ "embedding_dim = 100 # Embedding boyutu\n",
40
+ "max_len = X.shape[1] # Veri uzunluğu (her kelimenin vektör boyutu)\n",
41
+ "num_classes = len(np.unique(y)) # Sınıf sayısı (True/False)\n",
42
+ "\n",
43
+ "# LSTM Modeli oluşturma\n",
44
+ "model = Sequential([\n",
45
+ " Embedding(input_dim=len(X.columns) + 1, # Girdi boyutu, toplam kelime sayısı\n",
46
+ " output_dim=embedding_dim, # Embedding vektör boyutu\n",
47
+ " input_length=max_len), # Girdi uzunluğu\n",
48
+ " Dropout(0.2), # Aşırı öğrenmeyi engellemek için dropout\n",
49
+ " LSTM(128, activation='tanh'), # LSTM katmanı\n",
50
+ " Dense(64, activation='relu'), # Tam bağlı katman\n",
51
+ " Dense(num_classes, activation='softmax') # Çıkış katmanı (binary classification için softmax)\n",
52
+ "])\n",
53
+ "\n",
54
+ "# Modeli derleme\n",
55
+ "model.compile(loss='sparse_categorical_crossentropy', optimizer=Adam(), metrics=['accuracy'])\n",
56
+ "\n",
57
+ "# Modelin özetini görüntüleme\n",
58
+ "model.summary()\n",
59
+ "\n",
60
+ "# Modeli eğitme\n",
61
+ "history = model.fit(X_train, y_train, epochs=15, batch_size=64, validation_data=(X_test, y_test))\n",
62
+ "\n",
63
+ "# Eğitim ve doğrulama doğruluğunu görselleştirme\n",
64
+ "plt.plot(history.history['accuracy'], label='Eğitim Doğruluğu')\n",
65
+ "plt.plot(history.history['val_accuracy'], label='Doğrulama Doğruluğu')\n",
66
+ "plt.title('Model Doğruluğu')\n",
67
+ "plt.xlabel('Epok')\n",
68
+ "plt.ylabel('Doğruluk')\n",
69
+ "plt.legend(loc='best')\n",
70
+ "plt.show()\n",
71
+ "\n",
72
+ "# Test verisi üzerinde tahmin yapma\n",
73
+ "y_pred = model.predict(X_test)\n",
74
+ "y_pred_class = np.argmax(y_pred, axis=1) # Softmax çıktısından en yüksek sınıfı seçiyoruz\n",
75
+ "\n",
76
+ "# Accuracy\n",
77
+ "accuracy = accuracy_score(y_test, y_pred_class)\n",
78
+ "print(f'Accuracy: {accuracy:.4f}')\n",
79
+ "\n",
80
+ "# Confusion Matrix\n",
81
+ "cm = confusion_matrix(y_test, y_pred_class)\n",
82
+ "print('Confusion Matrix:')\n",
83
+ "print(cm)\n",
84
+ "\n",
85
+ "# Classification Report (Precision, Recall, F1-score)\n",
86
+ "class_report = classification_report(y_test, y_pred_class)\n",
87
+ "print('Classification Report:')\n",
88
+ "print(class_report)\n",
89
+ "\n",
90
+ "# ROC-AUC\n",
91
+ "roc_auc = roc_auc_score(y_test, model.predict(X_test)[:, 1]) # Binary classification için\n",
92
+ "print(f'ROC-AUC: {roc_auc:.4f}')\n",
93
+ "\n",
94
+ "# ROC Curve\n",
95
+ "fpr, tpr, thresholds = roc_curve(y_test, model.predict(X_test)[:, 1])\n",
96
+ "plt.figure()\n",
97
+ "plt.plot(fpr, tpr, color='blue', label=f'ROC curve (area = {roc_auc:.2f})')\n",
98
+ "plt.plot([0, 1], [0, 1], color='gray', linestyle='--')\n",
99
+ "plt.xlabel('False Positive Rate')\n",
100
+ "plt.ylabel('True Positive Rate')\n",
101
+ "plt.title('Receiver Operating Characteristic (ROC) Curve')\n",
102
+ "plt.legend(loc='lower right')\n",
103
+ "plt.show()\n",
104
+ "\n",
105
+ "\n"
106
+ ]
107
+ }
108
+ ],
109
+ "metadata": {
110
+ "language_info": {
111
+ "name": "python"
112
+ }
113
+ },
114
+ "nbformat": 4,
115
+ "nbformat_minor": 2
116
+ }
neural_networks.ipynb ADDED
@@ -0,0 +1,139 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cells": [
3
+ {
4
+ "cell_type": "code",
5
+ "execution_count": 1,
6
+ "metadata": {},
7
+ "outputs": [
8
+ {
9
+ "name": "stdout",
10
+ "output_type": "stream",
11
+ "text": [
12
+ "Neural Network Accuracy: 0.89\n",
13
+ "\n",
14
+ "Classification Report:\n",
15
+ " precision recall f1-score support\n",
16
+ "\n",
17
+ " 0 0.89 0.89 0.89 1301\n",
18
+ " 1 0.89 0.88 0.88 1244\n",
19
+ "\n",
20
+ " accuracy 0.89 2545\n",
21
+ " macro avg 0.89 0.89 0.89 2545\n",
22
+ "weighted avg 0.89 0.89 0.89 2545\n",
23
+ "\n"
24
+ ]
25
+ },
26
+ {
27
+ "data": {
28
+ "image/png": "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",
29
+ "text/plain": [
30
+ "<Figure size 800x600 with 2 Axes>"
31
+ ]
32
+ },
33
+ "metadata": {},
34
+ "output_type": "display_data"
35
+ },
36
+ {
37
+ "data": {
38
+ "image/png": "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",
39
+ "text/plain": [
40
+ "<Figure size 800x600 with 1 Axes>"
41
+ ]
42
+ },
43
+ "metadata": {},
44
+ "output_type": "display_data"
45
+ },
46
+ {
47
+ "data": {
48
+ "text/plain": [
49
+ "['neural_network_model.pkl']"
50
+ ]
51
+ },
52
+ "execution_count": 1,
53
+ "metadata": {},
54
+ "output_type": "execute_result"
55
+ }
56
+ ],
57
+ "source": [
58
+ "import pandas as pd \n",
59
+ "from sklearn.model_selection import train_test_split\n",
60
+ "from sklearn.metrics import accuracy_score, classification_report, confusion_matrix, roc_auc_score, roc_curve\n",
61
+ "from sklearn.neural_network import MLPClassifier\n",
62
+ "import seaborn as sns\n",
63
+ "import matplotlib.pyplot as plt\n",
64
+ "import joblib \n",
65
+ "\n",
66
+ "# Veriyi yükleme\n",
67
+ "data = pd.read_csv('data_vectorized.csv')\n",
68
+ "\n",
69
+ "# Veriyi ayırma (özellikler ve etiketler)\n",
70
+ "X = data.drop(columns=['Label']) # Etiket hariç tüm sütunlar\n",
71
+ "y = data['Label'] # Etiket\n",
72
+ "\n",
73
+ "# Veriyi eğitim ve test olarak ayırma\n",
74
+ "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n",
75
+ "\n",
76
+ "# Neural Network modelini oluşturma ve eğitim\n",
77
+ "nn_classifier = MLPClassifier(hidden_layer_sizes=(100,), max_iter=1000, random_state=42)\n",
78
+ "nn_classifier.fit(X_train, y_train)\n",
79
+ "\n",
80
+ "# Test seti üzerinde tahmin yapma\n",
81
+ "y_pred = nn_classifier.predict(X_test)\n",
82
+ "\n",
83
+ "# Doğruluk hesaplama\n",
84
+ "accuracy = accuracy_score(y_test, y_pred)\n",
85
+ "print(f\"Neural Network Accuracy: {accuracy:.2f}\")\n",
86
+ "\n",
87
+ "# Precision, Recall, F1-Score\n",
88
+ "print(\"\\nClassification Report:\")\n",
89
+ "print(classification_report(y_test, y_pred))\n",
90
+ "\n",
91
+ "# Confusion Matrix\n",
92
+ "conf_matrix = confusion_matrix(y_test, y_pred)\n",
93
+ "plt.figure(figsize=(8, 6))\n",
94
+ "sns.heatmap(conf_matrix, annot=True, fmt='d', cmap='Blues', xticklabels=[\"Class 0\", \"Class 1\"], yticklabels=[\"Class 0\", \"Class 1\"])\n",
95
+ "plt.title(\"Neural Network - Confusion Matrix\")\n",
96
+ "plt.xlabel(\"Predicted\")\n",
97
+ "plt.ylabel(\"Actual\")\n",
98
+ "plt.show()\n",
99
+ "\n",
100
+ "# ROC-AUC Hesaplama ve Grafik\n",
101
+ "if len(set(y_test)) == 2: # İkili sınıflandırma kontrolü\n",
102
+ " roc_auc = roc_auc_score(y_test, nn_classifier.predict_proba(X_test)[:, 1]) # Tahmin olasılıkları\n",
103
+ " fpr, tpr, thresholds = roc_curve(y_test, nn_classifier.predict_proba(X_test)[:, 1])\n",
104
+ " plt.figure(figsize=(8, 6))\n",
105
+ " plt.plot(fpr, tpr, label=f\"ROC-AUC: {roc_auc:.2f}\")\n",
106
+ " plt.plot([0, 1], [0, 1], 'k--', label=\"Random Guess\")\n",
107
+ " plt.xlabel(\"False Positive Rate\")\n",
108
+ " plt.ylabel(\"True Positive Rate\")\n",
109
+ " plt.title(\"Neural Network - ROC Curve\")\n",
110
+ " plt.legend()\n",
111
+ " plt.show()\n",
112
+ "\n",
113
+ "# Neural Network modelini kaydetme\n",
114
+ "joblib.dump(nn_classifier, 'neural_network_model.pkl') # Modeli kaydet\n"
115
+ ]
116
+ }
117
+ ],
118
+ "metadata": {
119
+ "kernelspec": {
120
+ "display_name": "Python 3",
121
+ "language": "python",
122
+ "name": "python3"
123
+ },
124
+ "language_info": {
125
+ "codemirror_mode": {
126
+ "name": "ipython",
127
+ "version": 3
128
+ },
129
+ "file_extension": ".py",
130
+ "mimetype": "text/x-python",
131
+ "name": "python",
132
+ "nbconvert_exporter": "python",
133
+ "pygments_lexer": "ipython3",
134
+ "version": "3.10.10"
135
+ }
136
+ },
137
+ "nbformat": 4,
138
+ "nbformat_minor": 2
139
+ }
pre_processing.ipynb ADDED
@@ -0,0 +1,363 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cells": [
3
+ {
4
+ "cell_type": "code",
5
+ "execution_count": 1,
6
+ "metadata": {},
7
+ "outputs": [
8
+ {
9
+ "name": "stdout",
10
+ "output_type": "stream",
11
+ "text": [
12
+ "Title 0\n",
13
+ "Label 0\n",
14
+ "dtype: int64\n"
15
+ ]
16
+ }
17
+ ],
18
+ "source": [
19
+ "import pandas as pd\n",
20
+ "\n",
21
+ "df = pd.read_csv(\"all_news.csv\")\n",
22
+ "\n",
23
+ "# Eksik değerleri kontrol ediyorum\n",
24
+ "print(df.isnull().sum())\n",
25
+ "\n",
26
+ "# Eksik değerleri dolduruyorum ya da kaldırıyorum\n",
27
+ "df = df.dropna()"
28
+ ]
29
+ },
30
+ {
31
+ "cell_type": "code",
32
+ "execution_count": 2,
33
+ "metadata": {},
34
+ "outputs": [
35
+ {
36
+ "name": "stdout",
37
+ "output_type": "stream",
38
+ "text": [
39
+ " Title Label\n",
40
+ "0 ABD Başkanı Trump ve destekçilerinin Rabia işa... YANLIŞ\n",
41
+ "1 Fotoğrafta Atatürk'ün güldüğü iddiası YANLIŞ\n",
42
+ "2 İsrail Başkonsolosluğu’nda ateş yakıldığını gö... YANLIŞ\n",
43
+ "3 Afişin 2022 Dünya Kupası’ndaki yasakları göste... YANLIŞ\n",
44
+ "4 İYİ Parti'de istifalar: Buğra Kavuncu ve Burak... DOĞRU\n"
45
+ ]
46
+ }
47
+ ],
48
+ "source": [
49
+ "# Label sütunu boş olan satırlarda hem Label hem de Title sütunlarını siliyoruz\n",
50
+ "df = df[df['Label'].notnull()]\n",
51
+ "\n",
52
+ "# Sonucu kontrol ediyoruz\n",
53
+ "print(df.head())"
54
+ ]
55
+ },
56
+ {
57
+ "cell_type": "code",
58
+ "execution_count": 3,
59
+ "metadata": {},
60
+ "outputs": [
61
+ {
62
+ "name": "stdout",
63
+ "output_type": "stream",
64
+ "text": [
65
+ "Empty DataFrame\n",
66
+ "Columns: [Title, Label]\n",
67
+ "Index: []\n"
68
+ ]
69
+ }
70
+ ],
71
+ "source": [
72
+ "df = pd.read_csv(\"all_news.csv\")\n",
73
+ "\n",
74
+ "# \"DOĞRU\" ve \"YANLIŞ\" dışındaki etiketleri filtreliyoruz\n",
75
+ "other_labels = df[~df['Label'].str.contains('DOĞRU|YANLIŞ', case=False, na=False)]\n",
76
+ "\n",
77
+ "# Bu satırları yazdırıyoruz, ilk 50 tanesini\n",
78
+ "print(other_labels[['Title', 'Label']].head(50))\n",
79
+ "# 'Label' sütunundaki boşlukları temizliyoruz\n",
80
+ "df['Label'] = df['Label'].str.strip()"
81
+ ]
82
+ },
83
+ {
84
+ "cell_type": "code",
85
+ "execution_count": 4,
86
+ "metadata": {},
87
+ "outputs": [],
88
+ "source": [
89
+ "#Metinlerin ön işlemi için kütüphaneler\n",
90
+ "import re\n",
91
+ "from nltk.corpus import stopwords\n",
92
+ "from nltk.stem import WordNetLemmatizer\n",
93
+ "import nltk\n",
94
+ "from sklearn.preprocessing import LabelEncoder"
95
+ ]
96
+ },
97
+ {
98
+ "cell_type": "code",
99
+ "execution_count": 5,
100
+ "metadata": {},
101
+ "outputs": [
102
+ {
103
+ "name": "stderr",
104
+ "output_type": "stream",
105
+ "text": [
106
+ "[nltk_data] Downloading package stopwords to C:\\Users\\Şerife\n",
107
+ "[nltk_data] Gökdaş\\AppData\\Roaming\\nltk_data...\n",
108
+ "[nltk_data] Package stopwords is already up-to-date!\n",
109
+ "[nltk_data] Downloading package wordnet to C:\\Users\\Şerife\n",
110
+ "[nltk_data] Gökdaş\\AppData\\Roaming\\nltk_data...\n",
111
+ "[nltk_data] Package wordnet is already up-to-date!\n"
112
+ ]
113
+ }
114
+ ],
115
+ "source": [
116
+ "nltk.download('stopwords')\n",
117
+ "nltk.download('wordnet')\n",
118
+ "# Stopwords ve lemmatizer tanımla\n",
119
+ "STOPWORDS = set(stopwords.words('turkish'))\n",
120
+ "STOPWORDS.add('mi') \n",
121
+ "lemmatizer = WordNetLemmatizer()"
122
+ ]
123
+ },
124
+ {
125
+ "cell_type": "code",
126
+ "execution_count": 6,
127
+ "metadata": {},
128
+ "outputs": [],
129
+ "source": [
130
+ "# Ön işleme fonksiyonu\n",
131
+ "def preprocess_text(text):\n",
132
+ " # Küçük harfe çevir\n",
133
+ " text = text.lower()\n",
134
+ " # Noktalama işaretlerini kaldır\n",
135
+ " text = re.sub(r'[^\\w\\s]', '', text)\n",
136
+ " # Stopwords kaldır ve lemmatization uygula\n",
137
+ " text = \" \".join([lemmatizer.lemmatize(word) for word in text.split() if word not in STOPWORDS])\n",
138
+ " return text"
139
+ ]
140
+ },
141
+ {
142
+ "cell_type": "code",
143
+ "execution_count": 7,
144
+ "metadata": {},
145
+ "outputs": [],
146
+ "source": [
147
+ "# CSV dosyasını oku\n",
148
+ "df = pd.read_csv('all_news.csv')\n",
149
+ "# Metinleri temizle\n",
150
+ "df['cleaned_text'] = df['Title'].apply(preprocess_text)"
151
+ ]
152
+ },
153
+ {
154
+ "cell_type": "code",
155
+ "execution_count": 8,
156
+ "metadata": {},
157
+ "outputs": [
158
+ {
159
+ "name": "stdout",
160
+ "output_type": "stream",
161
+ "text": [
162
+ "0 0\n",
163
+ "1 0\n",
164
+ "2 0\n",
165
+ "3 0\n",
166
+ "4 1\n",
167
+ "Name: Label, dtype: int64\n"
168
+ ]
169
+ }
170
+ ],
171
+ "source": [
172
+ "# Label sütununu sayısal değerlere çevirmek\n",
173
+ "label_encoder = LabelEncoder()\n",
174
+ "df['Label'] = label_encoder.fit_transform(df['Label'])\n",
175
+ "\n",
176
+ "# Dönüştürme sonrasında ters çevirme\n",
177
+ "df['Label'] = df['Label'].map({0: 1, 1: 0}) # 0 -> 1, 1 -> 0 \n",
178
+ "\n",
179
+ "# Etiket sınıfını kontrol et\n",
180
+ "print(df['Label'].head())\n",
181
+ "\n",
182
+ "# Temizlenmiş metin ve Label sütununu seçerek yeni bir dosya oluştur\n",
183
+ "df_cleaned = df[['cleaned_text', 'Label']]\n",
184
+ "df_cleaned.columns = ['Title', 'Label'] # Sütun isimlerini uygun şekilde düzenle"
185
+ ]
186
+ },
187
+ {
188
+ "cell_type": "code",
189
+ "execution_count": 9,
190
+ "metadata": {},
191
+ "outputs": [
192
+ {
193
+ "name": "stdout",
194
+ "output_type": "stream",
195
+ "text": [
196
+ " Title Label\n",
197
+ "0 abd başkanı trump destekçilerinin rabia işaret... 0\n",
198
+ "1 fotoğrafta atatürkün güldüğü iddiası 0\n",
199
+ "2 israil başkonsolosluğunda ateş yakıldığını gös... 0\n",
200
+ "3 afişin 2022 dünya kupasındaki yasakları göster... 0\n",
201
+ "4 iyi partide istifalar buğra kavuncu burak akbu... 1\n"
202
+ ]
203
+ }
204
+ ],
205
+ "source": [
206
+ "# Yeni CSV dosyasına sadece 'Title' ve 'Label' sütunlarını kaydet\n",
207
+ "df_cleaned.to_csv('data_all_news.csv', index=False, encoding='utf-8')\n",
208
+ "\n",
209
+ "# İlk birkaç satırı kontrol et\n",
210
+ "print(df_cleaned.head())"
211
+ ]
212
+ },
213
+ {
214
+ "cell_type": "code",
215
+ "execution_count": 10,
216
+ "metadata": {},
217
+ "outputs": [
218
+ {
219
+ "name": "stdout",
220
+ "output_type": "stream",
221
+ "text": [
222
+ "[('iddiası', 3692), ('bir', 1282), ('gösterdiği', 1166), ('gösteriyor', 989), ('video', 839), ('olduğu', 809), ('videonun', 618), ('fotoğrafın', 538), ('fotoğraf', 391), ('gerçek', 333), ('yeni', 297), ('ait', 290), ('güncel', 252), ('kişi', 248), ('doğru', 240), ('türkiye', 240), ('gösteren', 225), ('edilen', 224), ('iddia', 209), ('istanbul', 196)]\n"
223
+ ]
224
+ }
225
+ ],
226
+ "source": [
227
+ "#Title sayısallaştırmadan önce max_words değerime karar vermek için\n",
228
+ "import pandas as pd\n",
229
+ "\n",
230
+ "# CSV dosyasını yükleyin\n",
231
+ "df = pd.read_csv('data_all_news.csv')\n",
232
+ "\n",
233
+ "# Kelimelerin sıklığını incelemek\n",
234
+ "from collections import Counter\n",
235
+ "all_words = [word for text in df['Title'] for word in text.split()]\n",
236
+ "word_counts = Counter(all_words)\n",
237
+ "\n",
238
+ "# En yaygın 20 kelimeyi görmek\n",
239
+ "print(word_counts.most_common(20))\n",
240
+ "#çıktıya göre max_words 1500 olarak belirleyebilirim"
241
+ ]
242
+ },
243
+ {
244
+ "cell_type": "code",
245
+ "execution_count": 11,
246
+ "metadata": {},
247
+ "outputs": [
248
+ {
249
+ "name": "stdout",
250
+ "output_type": "stream",
251
+ "text": [
252
+ "500 kelime ile toplam kelime frekansının %38.58'i kapsanıyor.\n",
253
+ "1000 kelime ile toplam kelime frekansının %48.47'i kapsanıyor.\n",
254
+ "1500 kelime ile toplam kelime frekansının %55.19'i kapsanıyor.\n",
255
+ "2000 kelime ile toplam kelime frekansının %60.30'i kapsanıyor.\n"
256
+ ]
257
+ }
258
+ ],
259
+ "source": [
260
+ "import numpy as np\n",
261
+ "from collections import Counter\n",
262
+ "\n",
263
+ "# Kelime frekanslarını hesapla\n",
264
+ "all_words = [word for text in df['Title'] for word in text.split()]\n",
265
+ "word_counts = Counter(all_words)\n",
266
+ "\n",
267
+ "# Kelime frekanslarını sıralayıp birikimli toplamını al\n",
268
+ "word_freq = [count for _, count in word_counts.most_common()]\n",
269
+ "cumulative_freq = np.cumsum(word_freq)\n",
270
+ "total_words = sum(word_freq)\n",
271
+ "\n",
272
+ "# Belirli max_words için kapsanan yüzdeyi göster\n",
273
+ "for max_words in [500, 1000, 1500, 2000]:\n",
274
+ " coverage = (cumulative_freq[max_words-1] / total_words) * 100\n",
275
+ " print(f\"{max_words} kelime ile toplam kelime frekansının %{coverage:.2f}'i kapsanıyor.\")"
276
+ ]
277
+ },
278
+ {
279
+ "cell_type": "code",
280
+ "execution_count": 12,
281
+ "metadata": {},
282
+ "outputs": [
283
+ {
284
+ "name": "stdout",
285
+ "output_type": "stream",
286
+ "text": [
287
+ "count 12723.000000\n",
288
+ "mean 8.370824\n",
289
+ "std 2.708271\n",
290
+ "min 1.000000\n",
291
+ "25% 7.000000\n",
292
+ "50% 8.000000\n",
293
+ "75% 10.000000\n",
294
+ "max 25.000000\n",
295
+ "Name: title_length, dtype: float64\n"
296
+ ]
297
+ },
298
+ {
299
+ "data": {
300
+ "image/png": "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",
301
+ "text/plain": [
302
+ "<Figure size 640x480 with 1 Axes>"
303
+ ]
304
+ },
305
+ "metadata": {},
306
+ "output_type": "display_data"
307
+ }
308
+ ],
309
+ "source": [
310
+ "#Title sayısallaştırmadan önce max_len değerime karar vermek için\n",
311
+ "import pandas as pd\n",
312
+ "\n",
313
+ "# CSV dosyasını yükle\n",
314
+ "df = pd.read_csv(\"data_all_news.csv\")\n",
315
+ "\n",
316
+ "# Başlıkların uzunluklarını hesaplama (kelime sayısı)\n",
317
+ "df['title_length'] = df['Title'].apply(lambda x: len(x.split()))\n",
318
+ "\n",
319
+ "# Uzunlukların istatistiklerini görmek\n",
320
+ "print(df['title_length'].describe())\n",
321
+ "\n",
322
+ "# Başlık uzunluklarının dağılımını görmek için\n",
323
+ "import matplotlib.pyplot as plt\n",
324
+ "plt.hist(df['title_length'], bins=30, color='skyblue', edgecolor='black')\n",
325
+ "plt.title('Başlık Uzunluğu Dağılımı')\n",
326
+ "plt.xlabel('Başlık Uzunluğu (Kelime Sayısı)')\n",
327
+ "plt.ylabel('Frekans')\n",
328
+ "plt.show()\n",
329
+ "#çıktıya göre max_len 15 olarak belirleyebilirim"
330
+ ]
331
+ },
332
+ {
333
+ "cell_type": "code",
334
+ "execution_count": null,
335
+ "metadata": {},
336
+ "outputs": [],
337
+ "source": [
338
+ "#tokenizer_data.ipynb da tokenizer işlemleri yapıldı."
339
+ ]
340
+ }
341
+ ],
342
+ "metadata": {
343
+ "kernelspec": {
344
+ "display_name": "Python 3",
345
+ "language": "python",
346
+ "name": "python3"
347
+ },
348
+ "language_info": {
349
+ "codemirror_mode": {
350
+ "name": "ipython",
351
+ "version": 3
352
+ },
353
+ "file_extension": ".py",
354
+ "mimetype": "text/x-python",
355
+ "name": "python",
356
+ "nbconvert_exporter": "python",
357
+ "pygments_lexer": "ipython3",
358
+ "version": "3.10.10"
359
+ }
360
+ },
361
+ "nbformat": 4,
362
+ "nbformat_minor": 2
363
+ }
random_forest.ipynb ADDED
@@ -0,0 +1,141 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cells": [
3
+ {
4
+ "cell_type": "code",
5
+ "execution_count": 1,
6
+ "metadata": {},
7
+ "outputs": [
8
+ {
9
+ "name": "stdout",
10
+ "output_type": "stream",
11
+ "text": [
12
+ "Random Forest Accuracy: 0.86\n",
13
+ "\n",
14
+ "Classification Report:\n",
15
+ " precision recall f1-score support\n",
16
+ "\n",
17
+ " 0 0.86 0.88 0.87 1301\n",
18
+ " 1 0.87 0.84 0.86 1244\n",
19
+ "\n",
20
+ " accuracy 0.86 2545\n",
21
+ " macro avg 0.86 0.86 0.86 2545\n",
22
+ "weighted avg 0.86 0.86 0.86 2545\n",
23
+ "\n"
24
+ ]
25
+ },
26
+ {
27
+ "data": {
28
+ "image/png": "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",
29
+ "text/plain": [
30
+ "<Figure size 800x600 with 2 Axes>"
31
+ ]
32
+ },
33
+ "metadata": {},
34
+ "output_type": "display_data"
35
+ },
36
+ {
37
+ "data": {
38
+ "image/png": "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",
39
+ "text/plain": [
40
+ "<Figure size 800x600 with 1 Axes>"
41
+ ]
42
+ },
43
+ "metadata": {},
44
+ "output_type": "display_data"
45
+ },
46
+ {
47
+ "data": {
48
+ "text/plain": [
49
+ "['random_forest_model.pkl']"
50
+ ]
51
+ },
52
+ "execution_count": 1,
53
+ "metadata": {},
54
+ "output_type": "execute_result"
55
+ }
56
+ ],
57
+ "source": [
58
+ "import pandas as pd\n",
59
+ "from sklearn.model_selection import train_test_split\n",
60
+ "from sklearn.ensemble import RandomForestClassifier\n",
61
+ "from sklearn.metrics import accuracy_score, classification_report, confusion_matrix, roc_auc_score, roc_curve\n",
62
+ "import seaborn as sns\n",
63
+ "import matplotlib.pyplot as plt\n",
64
+ "import joblib \n",
65
+ "\n",
66
+ "# Veriyi yükleme\n",
67
+ "data = pd.read_csv('data_vectorized.csv')\n",
68
+ "\n",
69
+ "# Veriyi ayırma (özellikler ve etiketler)\n",
70
+ "X = data.drop(columns=['Label']) # Etiket hariç tüm sütunlar\n",
71
+ "y = data['Label'] # Etiket\n",
72
+ "\n",
73
+ "# Veriyi eğitim ve test olarak ayırma\n",
74
+ "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n",
75
+ "\n",
76
+ "# Random Forest modelini oluşturma ve eğitim\n",
77
+ "rf_classifier = RandomForestClassifier(n_estimators=100, random_state=42)\n",
78
+ "rf_classifier.fit(X_train, y_train)\n",
79
+ "\n",
80
+ "# Test seti üzerinde tahmin yapma\n",
81
+ "y_pred = rf_classifier.predict(X_test)\n",
82
+ "\n",
83
+ "# Doğruluk hesaplama\n",
84
+ "accuracy = accuracy_score(y_test, y_pred)\n",
85
+ "print(f\"Random Forest Accuracy: {accuracy:.2f}\")\n",
86
+ "\n",
87
+ "# Precision, Recall, F1-Score\n",
88
+ "print(\"\\nClassification Report:\")\n",
89
+ "print(classification_report(y_test, y_pred))\n",
90
+ "\n",
91
+ "# Confusion Matrix\n",
92
+ "conf_matrix = confusion_matrix(y_test, y_pred)\n",
93
+ "plt.figure(figsize=(8, 6))\n",
94
+ "sns.heatmap(conf_matrix, annot=True, fmt='d', cmap='Blues', xticklabels=[\"Class 0\", \"Class 1\"], yticklabels=[\"Class 0\", \"Class 1\"])\n",
95
+ "plt.title(\"Random Forest - Confusion Matrix\")\n",
96
+ "plt.xlabel(\"Predicted\")\n",
97
+ "plt.ylabel(\"Actual\")\n",
98
+ "plt.show()\n",
99
+ "\n",
100
+ "# ROC-AUC Hesaplama ve Grafik\n",
101
+ "if len(set(y_test)) == 2: # İkili sınıflandırma kontrolü\n",
102
+ " roc_auc = roc_auc_score(y_test, rf_classifier.predict_proba(X_test)[:, 1]) # Tahmin olasılıkları\n",
103
+ " fpr, tpr, thresholds = roc_curve(y_test, rf_classifier.predict_proba(X_test)[:, 1])\n",
104
+ " plt.figure(figsize=(8, 6))\n",
105
+ " plt.plot(fpr, tpr, label=f\"ROC-AUC: {roc_auc:.2f}\")\n",
106
+ " plt.plot([0, 1], [0, 1], 'k--', label=\"Random Guess\")\n",
107
+ " plt.xlabel(\"False Positive Rate\")\n",
108
+ " plt.ylabel(\"True Positive Rate\")\n",
109
+ " plt.title(\"Random Forest - ROC Curve\")\n",
110
+ " plt.legend()\n",
111
+ " plt.show()\n",
112
+ "\n",
113
+ " \n",
114
+ "\n",
115
+ "# Random Forest modelini kaydetme\n",
116
+ "joblib.dump(rf_classifier, 'random_forest_model.pkl') # Modeli kaydet\n"
117
+ ]
118
+ }
119
+ ],
120
+ "metadata": {
121
+ "kernelspec": {
122
+ "display_name": "Python 3",
123
+ "language": "python",
124
+ "name": "python3"
125
+ },
126
+ "language_info": {
127
+ "codemirror_mode": {
128
+ "name": "ipython",
129
+ "version": 3
130
+ },
131
+ "file_extension": ".py",
132
+ "mimetype": "text/x-python",
133
+ "name": "python",
134
+ "nbconvert_exporter": "python",
135
+ "pygments_lexer": "ipython3",
136
+ "version": "3.10.10"
137
+ }
138
+ },
139
+ "nbformat": 4,
140
+ "nbformat_minor": 2
141
+ }
requirements.txt ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ transformers
2
+ sentence-transformers
3
+ scikit-learn
4
+ numpy
5
+ streamlit
6
+ pandas
7
+ matplotlib
8
+ seaborn
9
+ tensorflow
10
+ joblib
11
+ xgboost
12
+ nltk
13
+ torch
14
+ tf-keras
support_vector_machine.ipynb ADDED
@@ -0,0 +1,139 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cells": [
3
+ {
4
+ "cell_type": "code",
5
+ "execution_count": 1,
6
+ "metadata": {},
7
+ "outputs": [
8
+ {
9
+ "name": "stdout",
10
+ "output_type": "stream",
11
+ "text": [
12
+ "SVM Accuracy: 0.87\n",
13
+ "\n",
14
+ "Classification Report:\n",
15
+ " precision recall f1-score support\n",
16
+ "\n",
17
+ " 0 0.90 0.84 0.87 1301\n",
18
+ " 1 0.85 0.90 0.87 1244\n",
19
+ "\n",
20
+ " accuracy 0.87 2545\n",
21
+ " macro avg 0.87 0.87 0.87 2545\n",
22
+ "weighted avg 0.87 0.87 0.87 2545\n",
23
+ "\n"
24
+ ]
25
+ },
26
+ {
27
+ "data": {
28
+ "image/png": "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",
29
+ "text/plain": [
30
+ "<Figure size 800x600 with 2 Axes>"
31
+ ]
32
+ },
33
+ "metadata": {},
34
+ "output_type": "display_data"
35
+ },
36
+ {
37
+ "data": {
38
+ "image/png": "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",
39
+ "text/plain": [
40
+ "<Figure size 800x600 with 1 Axes>"
41
+ ]
42
+ },
43
+ "metadata": {},
44
+ "output_type": "display_data"
45
+ },
46
+ {
47
+ "data": {
48
+ "text/plain": [
49
+ "['support_vector_machine_model.pkl']"
50
+ ]
51
+ },
52
+ "execution_count": 1,
53
+ "metadata": {},
54
+ "output_type": "execute_result"
55
+ }
56
+ ],
57
+ "source": [
58
+ "import pandas as pd\n",
59
+ "from sklearn.model_selection import train_test_split\n",
60
+ "from sklearn.svm import SVC\n",
61
+ "from sklearn.metrics import accuracy_score, classification_report, confusion_matrix, roc_auc_score, roc_curve\n",
62
+ "import seaborn as sns\n",
63
+ "import matplotlib.pyplot as plt\n",
64
+ "import joblib \n",
65
+ "\n",
66
+ "# Veriyi yükleme\n",
67
+ "data = pd.read_csv('data_vectorized.csv')\n",
68
+ "\n",
69
+ "# Veriyi ayırma (özellikler ve etiketler)\n",
70
+ "X = data.drop(columns=['Label']) # Etiket hariç tüm sütunlar\n",
71
+ "y = data['Label'] # Etiket\n",
72
+ "\n",
73
+ "# Veriyi eğitim ve test olarak ayırma\n",
74
+ "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n",
75
+ "\n",
76
+ "# SVM modelini oluşturma ve eğitim\n",
77
+ "svm_classifier = SVC(kernel='linear', probability=True, random_state=42) # 'linear' kernel kullanıyoruz\n",
78
+ "svm_classifier.fit(X_train, y_train)\n",
79
+ "\n",
80
+ "# Test seti üzerinde tahmin yapma\n",
81
+ "y_pred = svm_classifier.predict(X_test)\n",
82
+ "\n",
83
+ "# Doğruluk hesaplama\n",
84
+ "accuracy = accuracy_score(y_test, y_pred)\n",
85
+ "print(f\"SVM Accuracy: {accuracy:.2f}\")\n",
86
+ "\n",
87
+ "# Precision, Recall, F1-Score\n",
88
+ "print(\"\\nClassification Report:\")\n",
89
+ "print(classification_report(y_test, y_pred))\n",
90
+ "\n",
91
+ "# Confusion Matrix\n",
92
+ "conf_matrix = confusion_matrix(y_test, y_pred)\n",
93
+ "plt.figure(figsize=(8, 6))\n",
94
+ "sns.heatmap(conf_matrix, annot=True, fmt='d', cmap='Blues', xticklabels=[\"Class 0\", \"Class 1\"], yticklabels=[\"Class 0\", \"Class 1\"])\n",
95
+ "plt.title(\"SVM - Confusion Matrix\")\n",
96
+ "plt.xlabel(\"Predicted\")\n",
97
+ "plt.ylabel(\"Actual\")\n",
98
+ "plt.show()\n",
99
+ "\n",
100
+ "# ROC-AUC Hesaplama ve Grafik\n",
101
+ "if len(set(y_test)) == 2: # İkili sınıflandırma kontrolü\n",
102
+ " roc_auc = roc_auc_score(y_test, svm_classifier.predict_proba(X_test)[:, 1]) # Tahmin olasılıkları\n",
103
+ " fpr, tpr, thresholds = roc_curve(y_test, svm_classifier.predict_proba(X_test)[:, 1])\n",
104
+ " plt.figure(figsize=(8, 6))\n",
105
+ " plt.plot(fpr, tpr, label=f\"ROC-AUC: {roc_auc:.2f}\")\n",
106
+ " plt.plot([0, 1], [0, 1], 'k--', label=\"Random Guess\")\n",
107
+ " plt.xlabel(\"False Positive Rate\")\n",
108
+ " plt.ylabel(\"True Positive Rate\")\n",
109
+ " plt.title(\"SVM - ROC Curve\")\n",
110
+ " plt.legend()\n",
111
+ " plt.show()\n",
112
+ "\n",
113
+ "# SVM modelini kaydetme\n",
114
+ "joblib.dump(svm_classifier, 'support_vector_machine_model.pkl') # Modeli kaydet\n"
115
+ ]
116
+ }
117
+ ],
118
+ "metadata": {
119
+ "kernelspec": {
120
+ "display_name": "Python 3",
121
+ "language": "python",
122
+ "name": "python3"
123
+ },
124
+ "language_info": {
125
+ "codemirror_mode": {
126
+ "name": "ipython",
127
+ "version": 3
128
+ },
129
+ "file_extension": ".py",
130
+ "mimetype": "text/x-python",
131
+ "name": "python",
132
+ "nbconvert_exporter": "python",
133
+ "pygments_lexer": "ipython3",
134
+ "version": "3.10.10"
135
+ }
136
+ },
137
+ "nbformat": 4,
138
+ "nbformat_minor": 2
139
+ }
tokenized_processing.ipynb ADDED
@@ -0,0 +1,222 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cells": [
3
+ {
4
+ "cell_type": "code",
5
+ "execution_count": null,
6
+ "metadata": {},
7
+ "outputs": [],
8
+ "source": [
9
+ "#Terminalden huggingface-cli login komutuyla login oldum."
10
+ ]
11
+ },
12
+ {
13
+ "cell_type": "code",
14
+ "execution_count": 1,
15
+ "metadata": {},
16
+ "outputs": [
17
+ {
18
+ "name": "stderr",
19
+ "output_type": "stream",
20
+ "text": [
21
+ "c:\\Users\\Şerife GÖKDAŞ\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\tqdm\\auto.py:21: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
22
+ " from .autonotebook import tqdm as notebook_tqdm\n"
23
+ ]
24
+ },
25
+ {
26
+ "name": "stdout",
27
+ "output_type": "stream",
28
+ "text": [
29
+ "100 veri işlendi.\n",
30
+ "200 veri işlendi.\n",
31
+ "300 veri işlendi.\n",
32
+ "400 veri işlendi.\n",
33
+ "500 veri işlendi.\n",
34
+ "600 veri işlendi.\n",
35
+ "700 veri işlendi.\n",
36
+ "800 veri işlendi.\n",
37
+ "900 veri işlendi.\n",
38
+ "1000 veri işlendi.\n",
39
+ "1100 veri işlendi.\n",
40
+ "1200 veri işlendi.\n",
41
+ "1300 veri işlendi.\n",
42
+ "1400 veri işlendi.\n",
43
+ "1500 veri işlendi.\n",
44
+ "1600 veri işlendi.\n",
45
+ "1700 veri işlendi.\n",
46
+ "1800 veri işlendi.\n",
47
+ "1900 veri işlendi.\n",
48
+ "2000 veri işlendi.\n",
49
+ "2100 veri işlendi.\n",
50
+ "2200 veri işlendi.\n",
51
+ "2300 veri işlendi.\n",
52
+ "2400 veri işlendi.\n",
53
+ "2500 veri işlendi.\n",
54
+ "2600 veri işlendi.\n",
55
+ "2700 veri işlendi.\n",
56
+ "2800 veri işlendi.\n",
57
+ "2900 veri işlendi.\n",
58
+ "3000 veri işlendi.\n",
59
+ "3100 veri işlendi.\n",
60
+ "3200 veri işlendi.\n",
61
+ "3300 veri işlendi.\n",
62
+ "3400 veri işlendi.\n",
63
+ "3500 veri işlendi.\n",
64
+ "3600 veri işlendi.\n",
65
+ "3700 veri işlendi.\n",
66
+ "3800 veri işlendi.\n",
67
+ "3900 veri işlendi.\n",
68
+ "4000 veri işlendi.\n",
69
+ "4100 veri işlendi.\n",
70
+ "4200 veri işlendi.\n",
71
+ "4300 veri işlendi.\n",
72
+ "4400 veri işlendi.\n",
73
+ "4500 veri işlendi.\n",
74
+ "4600 veri işlendi.\n",
75
+ "4700 veri işlendi.\n",
76
+ "4800 veri işlendi.\n",
77
+ "4900 veri işlendi.\n",
78
+ "5000 veri işlendi.\n",
79
+ "5100 veri işlendi.\n",
80
+ "5200 veri işlendi.\n",
81
+ "5300 veri işlendi.\n",
82
+ "5400 veri işlendi.\n",
83
+ "5500 veri işlendi.\n",
84
+ "5600 veri işlendi.\n",
85
+ "5700 veri işlendi.\n",
86
+ "5800 veri işlendi.\n",
87
+ "5900 veri işlendi.\n",
88
+ "6000 veri işlendi.\n",
89
+ "6100 veri işlendi.\n",
90
+ "6200 veri işlendi.\n",
91
+ "6300 veri işlendi.\n",
92
+ "6400 veri işlendi.\n",
93
+ "6500 veri işlendi.\n",
94
+ "6600 veri işlendi.\n",
95
+ "6700 veri işlendi.\n",
96
+ "6800 veri işlendi.\n",
97
+ "6900 veri işlendi.\n",
98
+ "7000 veri işlendi.\n",
99
+ "7100 veri işlendi.\n",
100
+ "7200 veri işlendi.\n",
101
+ "7300 veri işlendi.\n",
102
+ "7400 veri işlendi.\n",
103
+ "7500 veri işlendi.\n",
104
+ "7600 veri işlendi.\n",
105
+ "7700 veri işlendi.\n",
106
+ "7800 veri işlendi.\n",
107
+ "7900 veri işlendi.\n",
108
+ "8000 veri işlendi.\n",
109
+ "8100 veri işlendi.\n",
110
+ "8200 veri işlendi.\n",
111
+ "8300 veri işlendi.\n",
112
+ "8400 veri işlendi.\n",
113
+ "8500 veri işlendi.\n",
114
+ "8600 veri işlendi.\n",
115
+ "8700 veri işlendi.\n",
116
+ "8800 veri işlendi.\n",
117
+ "8900 veri işlendi.\n",
118
+ "9000 veri işlendi.\n",
119
+ "9100 veri işlendi.\n",
120
+ "9200 veri işlendi.\n",
121
+ "9300 veri işlendi.\n",
122
+ "9400 veri işlendi.\n",
123
+ "9500 veri işlendi.\n",
124
+ "9600 veri işlendi.\n",
125
+ "9700 veri işlendi.\n",
126
+ "9800 veri işlendi.\n",
127
+ "9900 veri işlendi.\n",
128
+ "10000 veri işlendi.\n",
129
+ "10100 veri işlendi.\n",
130
+ "10200 veri işlendi.\n",
131
+ "10300 veri işlendi.\n",
132
+ "10400 veri işlendi.\n",
133
+ "10500 veri işlendi.\n",
134
+ "10600 veri işlendi.\n",
135
+ "10700 veri işlendi.\n",
136
+ "10800 veri işlendi.\n",
137
+ "10900 veri işlendi.\n",
138
+ "11000 veri işlendi.\n",
139
+ "11100 veri işlendi.\n",
140
+ "11200 veri işlendi.\n",
141
+ "11300 veri işlendi.\n",
142
+ "11400 veri işlendi.\n",
143
+ "11500 veri işlendi.\n",
144
+ "11600 veri işlendi.\n",
145
+ "11700 veri işlendi.\n",
146
+ "11800 veri işlendi.\n",
147
+ "11900 veri işlendi.\n",
148
+ "12000 veri işlendi.\n",
149
+ "12100 veri işlendi.\n",
150
+ "12200 veri işlendi.\n",
151
+ "12300 veri işlendi.\n",
152
+ "12400 veri işlendi.\n",
153
+ "12500 veri işlendi.\n",
154
+ "12600 veri işlendi.\n",
155
+ "12700 veri işlendi.\n",
156
+ "Tokenize işlemi tamamlandı ve yeni CSV dosyasına kaydedildi.\n"
157
+ ]
158
+ }
159
+ ],
160
+ "source": [
161
+ "import pandas as pd\n",
162
+ "from transformers import AutoTokenizer\n",
163
+ "\n",
164
+ "# Tokenizer'ı başlatıyoruz\n",
165
+ "tokenizer = AutoTokenizer.from_pretrained(\"alibayram/tr_tokenizer\", use_fast=True)\n",
166
+ "\n",
167
+ "# CSV dosyasını okuyalım\n",
168
+ "df = pd.read_csv(\"data_all_news.csv\")\n",
169
+ "\n",
170
+ "# Başlıkları (Title) ve etiketleri (Label) alıyoruz\n",
171
+ "titles = df['Title'].tolist() # 'Title' sütununu al\n",
172
+ "labels = df['Label'].tolist() # 'Label' sütununu al\n",
173
+ "\n",
174
+ "# Tokenize edilmiş başlıklar ve etiketleri saklamak için liste oluşturuyoruz\n",
175
+ "tokenized_titles = []\n",
176
+ "encoded_labels = []\n",
177
+ "\n",
178
+ "# Tokenize işlemine başlıyoruz\n",
179
+ "for i, (title, label) in enumerate(zip(titles, labels)):\n",
180
+ " tokens = tokenizer.tokenize(title) # Başlığı tokenize ediyoruz\n",
181
+ " tokenized_titles.append(tokens) # Tokenize edilmiş başlıkları listeye ekliyoruz\n",
182
+ " encoded_labels.append(label) # Etiketi listeye ekliyoruz\n",
183
+ "\n",
184
+ " # Her 100. veri işlendiğinde ekrana yazdırıyoruz\n",
185
+ " if (i + 1) % 100 == 0:\n",
186
+ " print(f\"{i + 1} veri işlendi.\")\n",
187
+ "\n",
188
+ "# Tokenize edilmiş başlıklar ve etiketlerle DataFrame oluşturuyoruz\n",
189
+ "tokenized_df = pd.DataFrame({\n",
190
+ " 'Title': tokenized_titles, # Tokenize edilmiş başlıklar\n",
191
+ " 'Label': encoded_labels # Etiketler\n",
192
+ "})\n",
193
+ "\n",
194
+ "# CSV dosyasına kaydediyoruz\n",
195
+ "tokenized_df.to_csv('data_tokenized.csv', index=False)\n",
196
+ "\n",
197
+ "print(\"Tokenize işlemi tamamlandı ve yeni CSV dosyasına kaydedildi.\")\n"
198
+ ]
199
+ }
200
+ ],
201
+ "metadata": {
202
+ "kernelspec": {
203
+ "display_name": "Python 3",
204
+ "language": "python",
205
+ "name": "python3"
206
+ },
207
+ "language_info": {
208
+ "codemirror_mode": {
209
+ "name": "ipython",
210
+ "version": 3
211
+ },
212
+ "file_extension": ".py",
213
+ "mimetype": "text/x-python",
214
+ "name": "python",
215
+ "nbconvert_exporter": "python",
216
+ "pygments_lexer": "ipython3",
217
+ "version": "3.10.10"
218
+ }
219
+ },
220
+ "nbformat": 4,
221
+ "nbformat_minor": 2
222
+ }
vector_processing.ipynb ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cells": [
3
+ {
4
+ "cell_type": "code",
5
+ "execution_count": 1,
6
+ "metadata": {},
7
+ "outputs": [
8
+ {
9
+ "name": "stdout",
10
+ "output_type": "stream",
11
+ "text": [
12
+ "False\n",
13
+ "0\n"
14
+ ]
15
+ }
16
+ ],
17
+ "source": [
18
+ "import torch\n",
19
+ "print(torch.cuda.is_available()) # True dönerse GPU kullanılabilir.\n",
20
+ "print(torch.cuda.device_count()) # GPU sayısını gösterir.\n"
21
+ ]
22
+ },
23
+ {
24
+ "cell_type": "code",
25
+ "execution_count": 3,
26
+ "metadata": {},
27
+ "outputs": [
28
+ {
29
+ "name": "stdout",
30
+ "output_type": "stream",
31
+ "text": [
32
+ "Vektörleştirme işlemi tamamlandı ve sonuçlar 'data_vectorized.csv' dosyasına kaydedildi.\n"
33
+ ]
34
+ }
35
+ ],
36
+ "source": [
37
+ "import pandas as pd\n",
38
+ "from sentence_transformers import SentenceTransformer\n",
39
+ "\n",
40
+ "# SBERT modelini yükleyelim\n",
41
+ "model = SentenceTransformer('sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2')\n",
42
+ "\n",
43
+ "# Tokenize edilmiş veriyi okuyalım\n",
44
+ "df = pd.read_csv(\"data_tokenized.csv\")\n",
45
+ "\n",
46
+ "# Tokenize edilmiş başlıkları alalım\n",
47
+ "tokenized_titles = df['Title'].tolist()\n",
48
+ "\n",
49
+ "# Tokenized listeyi string'e dönüştürelim\n",
50
+ "titles_as_strings = [' '.join(eval(tokens)) for tokens in tokenized_titles]\n",
51
+ "\n",
52
+ "# Başlıkları vektörleştirelim\n",
53
+ "vectorized_titles = model.encode(titles_as_strings)\n",
54
+ "\n",
55
+ "# Vektörleri DataFrame'e dönüştürelim\n",
56
+ "vectorized_df = pd.DataFrame(vectorized_titles)\n",
57
+ "vectorized_df['Label'] = df['Label'] # Etiketleri ekleyelim\n",
58
+ "\n",
59
+ "# DataFrame'i kaydedelim\n",
60
+ "vectorized_df.to_csv(\"data_vectorized.csv\", index=False)\n",
61
+ "\n",
62
+ "print(\"Vektörleştirme işlemi tamamlandı ve sonuçlar 'data_vectorized.csv' dosyasına kaydedildi.\")\n"
63
+ ]
64
+ }
65
+ ],
66
+ "metadata": {
67
+ "kernelspec": {
68
+ "display_name": "Python 3",
69
+ "language": "python",
70
+ "name": "python3"
71
+ },
72
+ "language_info": {
73
+ "codemirror_mode": {
74
+ "name": "ipython",
75
+ "version": 3
76
+ },
77
+ "file_extension": ".py",
78
+ "mimetype": "text/x-python",
79
+ "name": "python",
80
+ "nbconvert_exporter": "python",
81
+ "pygments_lexer": "ipython3",
82
+ "version": "3.10.10"
83
+ }
84
+ },
85
+ "nbformat": 4,
86
+ "nbformat_minor": 2
87
+ }
xg_boost.ipynb ADDED
@@ -0,0 +1,149 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cells": [
3
+ {
4
+ "cell_type": "code",
5
+ "execution_count": 1,
6
+ "metadata": {},
7
+ "outputs": [
8
+ {
9
+ "name": "stderr",
10
+ "output_type": "stream",
11
+ "text": [
12
+ "c:\\Users\\Şerife GÖKDAŞ\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\xgboost\\core.py:158: UserWarning: [17:39:51] WARNING: C:\\buildkite-agent\\builds\\buildkite-windows-cpu-autoscaling-group-i-0c55ff5f71b100e98-1\\xgboost\\xgboost-ci-windows\\src\\learner.cc:740: \n",
13
+ "Parameters: { \"use_label_encoder\" } are not used.\n",
14
+ "\n",
15
+ " warnings.warn(smsg, UserWarning)\n"
16
+ ]
17
+ },
18
+ {
19
+ "name": "stdout",
20
+ "output_type": "stream",
21
+ "text": [
22
+ "XGBoost Accuracy: 0.87\n",
23
+ "\n",
24
+ "Classification Report:\n",
25
+ " precision recall f1-score support\n",
26
+ "\n",
27
+ " 0 0.89 0.86 0.88 1301\n",
28
+ " 1 0.86 0.89 0.87 1244\n",
29
+ "\n",
30
+ " accuracy 0.87 2545\n",
31
+ " macro avg 0.87 0.87 0.87 2545\n",
32
+ "weighted avg 0.87 0.87 0.87 2545\n",
33
+ "\n"
34
+ ]
35
+ },
36
+ {
37
+ "data": {
38
+ "image/png": "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",
39
+ "text/plain": [
40
+ "<Figure size 800x600 with 2 Axes>"
41
+ ]
42
+ },
43
+ "metadata": {},
44
+ "output_type": "display_data"
45
+ },
46
+ {
47
+ "data": {
48
+ "image/png": "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",
49
+ "text/plain": [
50
+ "<Figure size 800x600 with 1 Axes>"
51
+ ]
52
+ },
53
+ "metadata": {},
54
+ "output_type": "display_data"
55
+ },
56
+ {
57
+ "data": {
58
+ "text/plain": [
59
+ "['xgboost_model.pkl']"
60
+ ]
61
+ },
62
+ "execution_count": 1,
63
+ "metadata": {},
64
+ "output_type": "execute_result"
65
+ }
66
+ ],
67
+ "source": [
68
+ "import pandas as pd\n",
69
+ "from sklearn.model_selection import train_test_split\n",
70
+ "from sklearn.metrics import accuracy_score, classification_report, confusion_matrix, roc_auc_score, roc_curve\n",
71
+ "import xgboost as xgb\n",
72
+ "import seaborn as sns\n",
73
+ "import matplotlib.pyplot as plt\n",
74
+ "import joblib \n",
75
+ "\n",
76
+ "# Veriyi yükleme\n",
77
+ "data = pd.read_csv('data_vectorized.csv')\n",
78
+ "\n",
79
+ "# Veriyi ayırma (özellikler ve etiketler)\n",
80
+ "X = data.drop(columns=['Label']) # Etiket hariç tüm sütunlar\n",
81
+ "y = data['Label'] # Etiket\n",
82
+ "\n",
83
+ "# Veriyi eğitim ve test olarak ayırma\n",
84
+ "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n",
85
+ "\n",
86
+ "# XGBoost modelini oluşturma ve eğitim\n",
87
+ "xgb_classifier = xgb.XGBClassifier(eval_metric='mlogloss', use_label_encoder=False, random_state=42)\n",
88
+ "xgb_classifier.fit(X_train, y_train)\n",
89
+ "\n",
90
+ "# Test seti üzerinde tahmin yapma\n",
91
+ "y_pred = xgb_classifier.predict(X_test)\n",
92
+ "\n",
93
+ "# Doğruluk hesaplama\n",
94
+ "accuracy = accuracy_score(y_test, y_pred)\n",
95
+ "print(f\"XGBoost Accuracy: {accuracy:.2f}\")\n",
96
+ "\n",
97
+ "# Precision, Recall, F1-Score\n",
98
+ "print(\"\\nClassification Report:\")\n",
99
+ "print(classification_report(y_test, y_pred))\n",
100
+ "\n",
101
+ "# Confusion Matrix\n",
102
+ "conf_matrix = confusion_matrix(y_test, y_pred)\n",
103
+ "plt.figure(figsize=(8, 6))\n",
104
+ "sns.heatmap(conf_matrix, annot=True, fmt='d', cmap='Blues', xticklabels=[\"Class 0\", \"Class 1\"], yticklabels=[\"Class 0\", \"Class 1\"])\n",
105
+ "plt.title(\"XGBoost - Confusion Matrix\")\n",
106
+ "plt.xlabel(\"Predicted\")\n",
107
+ "plt.ylabel(\"Actual\")\n",
108
+ "plt.show()\n",
109
+ "\n",
110
+ "# ROC-AUC Hesaplama ve Grafik\n",
111
+ "if len(set(y_test)) == 2: # İkili sınıflandırma kontrolü\n",
112
+ " roc_auc = roc_auc_score(y_test, xgb_classifier.predict_proba(X_test)[:, 1]) # Tahmin olasılıkları\n",
113
+ " fpr, tpr, thresholds = roc_curve(y_test, xgb_classifier.predict_proba(X_test)[:, 1])\n",
114
+ " plt.figure(figsize=(8, 6))\n",
115
+ " plt.plot(fpr, tpr, label=f\"ROC-AUC: {roc_auc:.2f}\")\n",
116
+ " plt.plot([0, 1], [0, 1], 'k--', label=\"Random Guess\")\n",
117
+ " plt.xlabel(\"False Positive Rate\")\n",
118
+ " plt.ylabel(\"True Positive Rate\")\n",
119
+ " plt.title(\"XGBoost - ROC Curve\")\n",
120
+ " plt.legend()\n",
121
+ " plt.show()\n",
122
+ "\n",
123
+ "# XGBoost modelini kaydetme\n",
124
+ "joblib.dump(xgb_classifier, 'xgboost_model.pkl') # Modeli kaydet\n"
125
+ ]
126
+ }
127
+ ],
128
+ "metadata": {
129
+ "kernelspec": {
130
+ "display_name": "Python 3",
131
+ "language": "python",
132
+ "name": "python3"
133
+ },
134
+ "language_info": {
135
+ "codemirror_mode": {
136
+ "name": "ipython",
137
+ "version": 3
138
+ },
139
+ "file_extension": ".py",
140
+ "mimetype": "text/x-python",
141
+ "name": "python",
142
+ "nbconvert_exporter": "python",
143
+ "pygments_lexer": "ipython3",
144
+ "version": "3.10.10"
145
+ }
146
+ },
147
+ "nbformat": 4,
148
+ "nbformat_minor": 2
149
+ }