File size: 34,613 Bytes
dfb3177 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 |
{
"cells": [
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Title 0\n",
"Label 0\n",
"dtype: int64\n"
]
}
],
"source": [
"import pandas as pd\n",
"\n",
"df = pd.read_csv(\"all_news.csv\")\n",
"\n",
"# Eksik değerleri kontrol ediyorum\n",
"print(df.isnull().sum())\n",
"\n",
"# Eksik değerleri dolduruyorum ya da kaldırıyorum\n",
"df = df.dropna()"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
" Title Label\n",
"0 ABD Başkanı Trump ve destekçilerinin Rabia işa... YANLIŞ\n",
"1 Fotoğrafta Atatürk'ün güldüğü iddiası YANLIŞ\n",
"2 İsrail Başkonsolosluğu’nda ateş yakıldığını gö... YANLIŞ\n",
"3 Afişin 2022 Dünya Kupası’ndaki yasakları göste... YANLIŞ\n",
"4 İYİ Parti'de istifalar: Buğra Kavuncu ve Burak... DOĞRU\n"
]
}
],
"source": [
"# Label sütunu boş olan satırlarda hem Label hem de Title sütunlarını siliyoruz\n",
"df = df[df['Label'].notnull()]\n",
"\n",
"# Sonucu kontrol ediyoruz\n",
"print(df.head())"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Empty DataFrame\n",
"Columns: [Title, Label]\n",
"Index: []\n"
]
}
],
"source": [
"df = pd.read_csv(\"all_news.csv\")\n",
"\n",
"# \"DOĞRU\" ve \"YANLIŞ\" dışındaki etiketleri filtreliyoruz\n",
"other_labels = df[~df['Label'].str.contains('DOĞRU|YANLIŞ', case=False, na=False)]\n",
"\n",
"# Bu satırları yazdırıyoruz, ilk 50 tanesini\n",
"print(other_labels[['Title', 'Label']].head(50))\n",
"# 'Label' sütunundaki boşlukları temizliyoruz\n",
"df['Label'] = df['Label'].str.strip()"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [],
"source": [
"#Metinlerin ön işlemi için kütüphaneler\n",
"import re\n",
"from nltk.corpus import stopwords\n",
"from nltk.stem import WordNetLemmatizer\n",
"import nltk\n",
"from sklearn.preprocessing import LabelEncoder"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"[nltk_data] Downloading package stopwords to C:\\Users\\Şerife\n",
"[nltk_data] Gökdaş\\AppData\\Roaming\\nltk_data...\n",
"[nltk_data] Package stopwords is already up-to-date!\n",
"[nltk_data] Downloading package wordnet to C:\\Users\\Şerife\n",
"[nltk_data] Gökdaş\\AppData\\Roaming\\nltk_data...\n",
"[nltk_data] Package wordnet is already up-to-date!\n"
]
}
],
"source": [
"nltk.download('stopwords')\n",
"nltk.download('wordnet')\n",
"# Stopwords ve lemmatizer tanımla\n",
"STOPWORDS = set(stopwords.words('turkish'))\n",
"STOPWORDS.add('mi') \n",
"lemmatizer = WordNetLemmatizer()"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {},
"outputs": [],
"source": [
"# Ön işleme fonksiyonu\n",
"def preprocess_text(text):\n",
" # Küçük harfe çevir\n",
" text = text.lower()\n",
" # Noktalama işaretlerini kaldır\n",
" text = re.sub(r'[^\\w\\s]', '', text)\n",
" # Stopwords kaldır ve lemmatization uygula\n",
" text = \" \".join([lemmatizer.lemmatize(word) for word in text.split() if word not in STOPWORDS])\n",
" return text"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {},
"outputs": [],
"source": [
"# CSV dosyasını oku\n",
"df = pd.read_csv('all_news.csv')\n",
"# Metinleri temizle\n",
"df['cleaned_text'] = df['Title'].apply(preprocess_text)"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"0 0\n",
"1 0\n",
"2 0\n",
"3 0\n",
"4 1\n",
"Name: Label, dtype: int64\n"
]
}
],
"source": [
"# Label sütununu sayısal değerlere çevirmek\n",
"label_encoder = LabelEncoder()\n",
"df['Label'] = label_encoder.fit_transform(df['Label'])\n",
"\n",
"# Dönüştürme sonrasında ters çevirme\n",
"df['Label'] = df['Label'].map({0: 1, 1: 0}) # 0 -> 1, 1 -> 0 \n",
"\n",
"# Etiket sınıfını kontrol et\n",
"print(df['Label'].head())\n",
"\n",
"# Temizlenmiş metin ve Label sütununu seçerek yeni bir dosya oluştur\n",
"df_cleaned = df[['cleaned_text', 'Label']]\n",
"df_cleaned.columns = ['Title', 'Label'] # Sütun isimlerini uygun şekilde düzenle"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
" Title Label\n",
"0 abd başkanı trump destekçilerinin rabia işaret... 0\n",
"1 fotoğrafta atatürkün güldüğü iddiası 0\n",
"2 israil başkonsolosluğunda ateş yakıldığını gös... 0\n",
"3 afişin 2022 dünya kupasındaki yasakları göster... 0\n",
"4 iyi partide istifalar buğra kavuncu burak akbu... 1\n"
]
}
],
"source": [
"# Yeni CSV dosyasına sadece 'Title' ve 'Label' sütunlarını kaydet\n",
"df_cleaned.to_csv('data_all_news.csv', index=False, encoding='utf-8')\n",
"\n",
"# İlk birkaç satırı kontrol et\n",
"print(df_cleaned.head())"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"[('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"
]
}
],
"source": [
"#Title sayısallaştırmadan önce max_words değerime karar vermek için\n",
"import pandas as pd\n",
"\n",
"# CSV dosyasını yükleyin\n",
"df = pd.read_csv('data_all_news.csv')\n",
"\n",
"# Kelimelerin sıklığını incelemek\n",
"from collections import Counter\n",
"all_words = [word for text in df['Title'] for word in text.split()]\n",
"word_counts = Counter(all_words)\n",
"\n",
"# En yaygın 20 kelimeyi görmek\n",
"print(word_counts.most_common(20))\n",
"#çıktıya göre max_words 1500 olarak belirleyebilirim"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"500 kelime ile toplam kelime frekansının %38.58'i kapsanıyor.\n",
"1000 kelime ile toplam kelime frekansının %48.47'i kapsanıyor.\n",
"1500 kelime ile toplam kelime frekansının %55.19'i kapsanıyor.\n",
"2000 kelime ile toplam kelime frekansının %60.30'i kapsanıyor.\n"
]
}
],
"source": [
"import numpy as np\n",
"from collections import Counter\n",
"\n",
"# Kelime frekanslarını hesapla\n",
"all_words = [word for text in df['Title'] for word in text.split()]\n",
"word_counts = Counter(all_words)\n",
"\n",
"# Kelime frekanslarını sıralayıp birikimli toplamını al\n",
"word_freq = [count for _, count in word_counts.most_common()]\n",
"cumulative_freq = np.cumsum(word_freq)\n",
"total_words = sum(word_freq)\n",
"\n",
"# Belirli max_words için kapsanan yüzdeyi göster\n",
"for max_words in [500, 1000, 1500, 2000]:\n",
" coverage = (cumulative_freq[max_words-1] / total_words) * 100\n",
" print(f\"{max_words} kelime ile toplam kelime frekansının %{coverage:.2f}'i kapsanıyor.\")"
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"count 12723.000000\n",
"mean 8.370824\n",
"std 2.708271\n",
"min 1.000000\n",
"25% 7.000000\n",
"50% 8.000000\n",
"75% 10.000000\n",
"max 25.000000\n",
"Name: title_length, dtype: float64\n"
]
},
{
"data": {
"image/png": "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",
"text/plain": [
"<Figure size 640x480 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"#Title sayısallaştırmadan önce max_len değerime karar vermek için\n",
"import pandas as pd\n",
"\n",
"# CSV dosyasını yükle\n",
"df = pd.read_csv(\"data_all_news.csv\")\n",
"\n",
"# Başlıkların uzunluklarını hesaplama (kelime sayısı)\n",
"df['title_length'] = df['Title'].apply(lambda x: len(x.split()))\n",
"\n",
"# Uzunlukların istatistiklerini görmek\n",
"print(df['title_length'].describe())\n",
"\n",
"# Başlık uzunluklarının dağılımını görmek için\n",
"import matplotlib.pyplot as plt\n",
"plt.hist(df['title_length'], bins=30, color='skyblue', edgecolor='black')\n",
"plt.title('Başlık Uzunluğu Dağılımı')\n",
"plt.xlabel('Başlık Uzunluğu (Kelime Sayısı)')\n",
"plt.ylabel('Frekans')\n",
"plt.show()\n",
"#çıktıya göre max_len 15 olarak belirleyebilirim"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"#tokenizer_data.ipynb da tokenizer işlemleri yapıldı."
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.10.10"
}
},
"nbformat": 4,
"nbformat_minor": 2
}
|