{"id":34968,"date":"2025-11-24T07:40:46","date_gmt":"2025-11-24T04:40:46","guid":{"rendered":"https:\/\/fatihsoysal.com\/blog\/?p=34968"},"modified":"2025-11-24T07:40:46","modified_gmt":"2025-11-24T04:40:46","slug":"python-3te-dil-verileriyle-calismak-dogal-dil-isleme-icin-nltk-kutuphanesi","status":"publish","type":"post","link":"https:\/\/fatihsoysal.com\/blog\/python-3te-dil-verileriyle-calismak-dogal-dil-isleme-icin-nltk-kutuphanesi\/","title":{"rendered":"Python 3’te Dil Verileriyle \u00c7al\u0131\u015fmak: Do\u011fal Dil \u0130\u015fleme \u0130\u00e7in NLTK K\u00fct\u00fcphanesi"},"content":{"rendered":"

<\/p>\n

Python 3’te Dil Verileriyle \u00c7al\u0131\u015fmak: Do\u011fal Dil \u0130\u015fleme \u0130\u00e7in NLTK K\u00fct\u00fcphanesi<\/h2>\n

Do\u011fal Dil \u0130\u015fleme (NLP), bilgisayarlar\u0131n insan dilini anlamas\u0131n\u0131, yorumlamas\u0131n\u0131 ve \u00fcretmesini sa\u011flayan yapay zeka alan\u0131n\u0131n \u00f6nemli bir dal\u0131d\u0131r. G\u00fcn\u00fcm\u00fcz\u00fcn veri odakl\u0131 d\u00fcnyas\u0131nda, metin verilerinin hacmi katlanarak artmaktad\u0131r ve bu verilerden anlaml\u0131 i\u00e7g\u00f6r\u00fcler elde etmek, i\u015fletmeler, ara\u015ft\u0131rmac\u0131lar ve geli\u015ftiriciler i\u00e7in kritik bir yetenek haline gelmi\u015ftir. Sosyal medya g\u00f6nderilerinden m\u00fc\u015fteri geri bildirimlerine, t\u0131bbi raporlardan hukuki belgelere kadar her alanda kar\u015f\u0131m\u0131za \u00e7\u0131kan dil verileri, do\u011fru ara\u00e7larla i\u015flendi\u011finde paha bi\u00e7ilmez de\u011ferler sunabilir.<\/p>\n

Python, basit s\u00f6zdizimi, geni\u015f k\u00fct\u00fcphane ekosistemi ve g\u00fc\u00e7l\u00fc topluluk deste\u011fi sayesinde NLP alan\u0131nda en \u00e7ok tercih edilen programlama dillerinden biridir. Bu ekosistemin temel ta\u015flar\u0131ndan biri de Natural Language Toolkit (NLTK) k\u00fct\u00fcphanesidir. NLTK, do\u011fal dil i\u015fleme g\u00f6revlerini ger\u00e7ekle\u015ftirmek i\u00e7in kapsaml\u0131 bir ara\u00e7 seti sunar. Metin \u00f6n i\u015fleme, s\u0131n\u0131fland\u0131rma, belirte\u00e7lere ay\u0131rma, k\u00f6k bulma, lemmatizasyon, s\u00f6zdizimsel ayr\u0131\u015ft\u0131rma ve anlamsal ak\u0131l y\u00fcr\u00fctme gibi bir\u00e7ok temel NLP g\u00f6revini kolayla\u015ft\u0131ran NLTK, \u00f6zellikle NLP’ye yeni ba\u015flayanlar ve akademik ara\u015ft\u0131rmac\u0131lar i\u00e7in vazge\u00e7ilmez bir kaynakt\u0131r.<\/p>\n

Bu makale, Python 3 kullanarak NLTK ile dil verileri \u00fczerinde nas\u0131l \u00e7al\u0131\u015f\u0131laca\u011f\u0131n\u0131 ad\u0131m ad\u0131m a\u00e7\u0131klamay\u0131 ama\u00e7lamaktad\u0131r. NLTK’n\u0131n kurulumundan ba\u015flayarak, metin \u00f6n i\u015fleme tekniklerine, temel analiz y\u00f6ntemlerine ve hatta basit dil modelleri olu\u015fturmaya kadar geni\u015f bir yelpazede pratik bilgiler ve kod \u00f6rnekleri sunulacakt\u0131r. Makalenin sonunda, okuyucular\u0131n NLTK’n\u0131n sundu\u011fu imkanlar\u0131 anlayarak kendi NLP projelerinde kullanabilecek temel yetkinliklere sahip olmalar\u0131 hedeflenmektedir.<\/p>\n

NLTK Kurulumu ve Temel Veri \u0130ndirme<\/h3>\n

NLTK’y\u0131 kullanmaya ba\u015flamadan \u00f6nce, Python 3 ortam\u0131n\u0131zda k\u00fct\u00fcphaneyi kurman\u0131z ve gerekli veri paketlerini indirmeniz gerekmektedir. Python’\u0131n kurulu oldu\u011funu varsayarak, NLTK’y\u0131 pip<\/code> arac\u0131l\u0131\u011f\u0131yla kolayca kurabilirsiniz.<\/p>\n

NLTK Kurulumu<\/h4>\n

Komut istemcinizi veya terminalinizi a\u00e7\u0131n ve a\u015fa\u011f\u0131daki komutu \u00e7al\u0131\u015ft\u0131r\u0131n:<\/p>\n

pip install nltk<\/code><\/pre>\n
NLTK Veri Paketlerini \u0130ndirme<\/h4>\n
NLTK, \u00e7o\u011fu i\u015flevi i\u00e7in harici veri setlerine ihtiya\u00e7 duyar. Bu veri setleri, durma kelimeleri listelerinden, POS (Part-of-Speech) etiketleyicilerine, WordNet s\u00f6zl\u00fc\u011f\u00fcnden, \u00e7e\u015fitli diller i\u00e7in belirte\u00e7lere ay\u0131rma modellerine kadar geni\u015f bir yelpazeyi kapsar. Bu verileri indirmek i\u00e7in Python beti\u011finizde nltk.download()<\/code> i\u015flevini kullanabilirsiniz:<\/p>\n
import nltk\nnltk.download()<\/code><\/pre>\n
Bu komutu \u00e7al\u0131\u015ft\u0131rd\u0131\u011f\u0131n\u0131zda, NLTK Downloader ad\u0131 verilen bir GUI penceresi a\u00e7\u0131lacakt\u0131r. Bu pencereden tek tek paketleri veya all<\/code> se\u00e7ene\u011fini i\u015faretleyerek t\u00fcm paketleri indirebilirsiniz. Genellikle, ilk ba\u015fta all<\/code> paketini indirmek, gelecekteki olas\u0131 ba\u011f\u0131ml\u0131l\u0131k sorunlar\u0131n\u0131 \u00f6nlemek i\u00e7in iyi bir yakla\u015f\u0131md\u0131r. Ancak, disk alan\u0131ndan tasarruf etmek ve sadece ihtiyac\u0131n\u0131z olanlar\u0131 indirmek isterseniz, belirli paketleri se\u00e7ebilirsiniz. \u00d6rne\u011fin, sadece punkt<\/code> (belirte\u00e7lere ay\u0131rma i\u00e7in) ve stopwords<\/code> (durma kelimeleri i\u00e7in) indirmek isterseniz:<\/p>\n
import nltk\nnltk.download('punkt')\nnltk.download('stopwords')\nnltk.download('wordnet')\nnltk.download('averaged_perceptron_tagger')<\/code><\/pre>\n
Metin \u00d6n \u0130\u015fleme Teknikleri<\/h3>\n
Do\u011fal dil verileri, genellikle g\u00fcr\u00fclt\u00fcl\u00fc, tutars\u0131z ve yap\u0131land\u0131r\u0131lmam\u0131\u015f haldedir. Bu verileri anlaml\u0131 analizler i\u00e7in uygun hale getirmek amac\u0131yla \u00e7e\u015fitli \u00f6n i\u015fleme ad\u0131mlar\u0131na tabi tutmak gerekir. NLTK, bu ad\u0131mlar\u0131n \u00e7o\u011funu kolayca ger\u00e7ekle\u015ftirmenizi sa\u011flar.<\/p>\n
Belirte\u00e7lere Ay\u0131rma (Tokenization)<\/h4>\n
Tokenization, metni daha k\u00fc\u00e7\u00fck birimlere, yani “belirte\u00e7lere” (token) ay\u0131rma i\u015flemidir. Bu belirte\u00e7ler genellikle kelimeler veya c\u00fcmleler olabilir.<\/p>\n
C\u00fcmle Belirte\u00e7lere Ay\u0131rma (Sentence Tokenization)<\/h5>\n
Bir metin blo\u011funu c\u00fcmlelerine ay\u0131rmak i\u00e7in sent_tokenize<\/code> fonksiyonunu kullan\u0131r\u0131z.<\/p>\n
from nltk.tokenize import sent_tokenize\n\ntext = \"NLTK, Python i\u00e7in g\u00fc\u00e7l\u00fc bir k\u00fct\u00fcphanedir. Do\u011fal Dil \u0130\u015fleme g\u00f6revlerinde yayg\u0131n olarak kullan\u0131l\u0131r. \u00d6\u011frenmesi ve uygulamas\u0131 olduk\u00e7a kolayd\u0131r.\"\nsentences = sent_tokenize(text)\nprint(\"C\u00fcmleler:\")\nfor s in sentences:\n    print(s)\n
\u00c7\u0131kt\u0131:<\/h2>\n
C\u00fcmleler:<\/h2>\n
NLTK, Python i\u00e7in g\u00fc\u00e7l\u00fc bir k\u00fct\u00fcphanedir.<\/h2>\n
Do\u011fal Dil \u0130\u015fleme g\u00f6revlerinde yayg\u0131n olarak kullan\u0131l\u0131r.<\/h2>\n
\u00d6\u011frenmesi ve uygulamas\u0131 olduk\u00e7a kolayd\u0131r.<\/code><\/pre>\n<\/h2>\n
Kelime Belirte\u00e7lere Ay\u0131rma (Word Tokenization)<\/h5>\n
Bir c\u00fcmleyi veya metin blo\u011funu kelimelerine ay\u0131rmak i\u00e7in word_tokenize<\/code> fonksiyonunu kullan\u0131r\u0131z.<\/p>\n
from nltk.tokenize import word_tokenize\n\nsentence = \"NLTK, do\u011fal dil i\u015fleme i\u00e7in harika bir ara\u00e7t\u0131r.\"\nwords = word_tokenize(sentence)\nprint(\"\\nKelime Belirte\u00e7leri:\")\nprint(words)\n
\u00c7\u0131kt\u0131:<\/h2>\n
Kelime Belirte\u00e7leri:<\/h2>\n
['NLTK', ',', 'do\u011fal', 'dil', 'i\u015fleme', 'i\u00e7in', 'harika', 'bir', 'ara\u00e7t\u0131r', '.']<\/code><\/pre>\n<\/h2>\n
T\u00fcrk\u00e7e metinler i\u00e7in de word_tokenize<\/code> genellikle iyi \u00e7al\u0131\u015f\u0131r, ancak \u00f6zel durumlar ve biti\u015fik kelimeler i\u00e7in daha geli\u015fmi\u015f modeller gerekebilir.<\/p>\n
Durma Kelimeleri (Stop Words)<\/h4>\n
Durma kelimeleri, bir dilde s\u0131k\u00e7a kullan\u0131lan ancak metnin anlam\u0131n\u0131 belirlemede \u00e7ok az katk\u0131s\u0131 olan kelimelerdir (\u00f6rn. “ve”, “bir”, “i\u00e7in”, “bu”). Bu kelimeleri metinden \u00e7\u0131karmak, analizlerin daha odakl\u0131 ve verimli olmas\u0131n\u0131 sa\u011flar.<\/p>\n
from nltk.corpus import stopwords\nfrom nltk.tokenize import word_tokenize\n\ntext = \"Bu, NLTK k\u00fct\u00fcphanesini kullanarak do\u011fal dil i\u015fleme \u00f6\u011frenmek i\u00e7in harika bir \u00f6rnektir.\"\nword_tokens = word_tokenize(text.lower()) # Metni k\u00fc\u00e7\u00fck harfe \u00e7evirmek \u00f6nemlidir\n\n
\u0130ngilizce durma kelimeleri<\/h2>\nenglish_stopwords = set(stopwords.words('english'))\nfiltered_english_words = [w for w in word_tokens if not w in english_stopwords]\nprint(\"\\n\u0130ngilizce Durma Kelimeleri \u00c7\u0131kar\u0131lm\u0131\u015f:\")\nprint(filtered_english_words)\n
\u00c7\u0131kt\u0131:<\/h2>\n
\u0130ngilizce Durma Kelimeleri \u00c7\u0131kar\u0131lm\u0131\u015f:<\/h2>\n
['bu', ',', 'nltk', 'k\u00fct\u00fcphanesini', 'kullanarak', 'do\u011fal', 'dil', 'i\u015fleme', '\u00f6\u011frenmek', 'i\u00e7in', 'harika', 'bir', '\u00f6rnektir', '.']<\/h2>\n\n
NLTK'da do\u011frudan T\u00fcrk\u00e7e durma kelimeleri paketi bulunmaz.<\/h2>\n
Ancak, harici kaynaklardan veya manuel olarak bir liste olu\u015fturulabilir.<\/h2>\n
\u00d6rnek bir T\u00fcrk\u00e7e durma kelimeleri listesi:<\/h2>\nturkish_stopwords = [\n    \"acaba\", \"ama\", \"asl\u0131nda\", \"az\", \"baz\u0131\", \"belki\", \"biri\", \"birka\u00e7\", \"bir\u015fey\", \"biz\", \"b\u00f6yle\", \"bunu\", \"\u00e7\u00fcnk\u00fc\",\n    \"\u00e7ok\", \"de\", \"daha\", \"da\", \"e\u011fer\", \"fakat\", \"gibi\", \"hem\", \"hi\u00e7\", \"i\u00e7in\", \"ile\", \"ise\", \"i\u015fte\", \"ka\u00e7\", \"kendi\",\n    \"ki\", \"kim\", \"mu\", \"m\u0131\", \"nas\u0131l\", \"ne\", \"neden\", \"nerde\", \"nereye\", \"niye\", \"o\", \"oysa\", \"\u00f6yle\", \"pek\", \"ra\u011fmen\",\n    \"sanki\", \"\u015fey\", \"siz\", \"\u015fu\", \"\u015funu\", \"tabii\", \"t\u00fcm\", \"ve\", \"veya\", \"ya\", \"yani\", \"yine\", \"\u00e7ok\", \"bu\", \"bir\"\n]\nfiltered_turkish_words = [w for w in word_tokens if not w in turkish_stopwords]\nprint(\"\\nT\u00fcrk\u00e7e Durma Kelimeleri \u00c7\u0131kar\u0131lm\u0131\u015f (\u00d6rnek Liste \u0130le):\")\nprint(filtered_turkish_words)\n
\u00c7\u0131kt\u0131:<\/h2>\n
T\u00fcrk\u00e7e Durma Kelimeleri \u00c7\u0131kar\u0131lm\u0131\u015f (\u00d6rnek Liste \u0130le):<\/h2>\n
[',', 'nltk', 'k\u00fct\u00fcphanesini', 'kullanarak', 'do\u011fal', 'dil', 'i\u015fleme', '\u00f6\u011frenmek', 'harika', '\u00f6rnektir', '.']<\/code><\/pre>\n<\/h2>\n
G\u00f6r\u00fcld\u00fc\u011f\u00fc gibi, T\u00fcrk\u00e7e i\u00e7in NLTK’n\u0131n kendi durma kelimeleri listesi olmamas\u0131na ra\u011fmen, manuel olarak veya ba\u015fka kaynaklardan edinilen listelerle bu i\u015flem yap\u0131labilir.<\/p>\n
B\u00fcy\u00fck\/K\u00fc\u00e7\u00fck Harf D\u00f6n\u00fc\u015f\u00fcm\u00fc (Case Folding)<\/h4>\n
Metni standart hale getirmek i\u00e7in genellikle t\u00fcm harfleri k\u00fc\u00e7\u00fck harfe d\u00f6n\u00fc\u015ft\u00fcr\u00fcl\u00fcr. Bu, “Elma” ve “elma” kelimelerinin ayn\u0131 kabul edilmesini sa\u011flar.<\/p>\n
text = \"Bu bir \u00d6rnek Metindir.\"\nlower_case_text = text.lower()\nprint(\"\\nK\u00fc\u00e7\u00fck Harfe D\u00f6n\u00fc\u015ft\u00fcr\u00fclm\u00fc\u015f Metin:\")\nprint(lower_case_text)\n
\u00c7\u0131kt\u0131:<\/h2>\n
K\u00fc\u00e7\u00fck Harfe D\u00f6n\u00fc\u015ft\u00fcr\u00fclm\u00fc\u015f Metin:<\/h2>\n
bu bir \u00f6rnek metindir.<\/code><\/pre>\n<\/h2>\n
Noktalama \u0130\u015faretleri ve Say\u0131lar\u0131n Temizlenmesi<\/h4>\n
Analiz hedefine ba\u011fl\u0131 olarak, noktalama i\u015faretleri ve say\u0131lar\u0131n metinden \u00e7\u0131kar\u0131lmas\u0131 gerekebilir. Bunun i\u00e7in Python’\u0131n re<\/code> (regular expression) mod\u00fcl\u00fc s\u0131k\u00e7a kullan\u0131l\u0131r.<\/p>\n
import re\nfrom nltk.tokenize import word_tokenize\n\ntext = \"NLTK ile 2023 y\u0131l\u0131nda NLP \u00f6\u011frenmek \u00e7ok keyifli! Bu konuda %100 eminim.\"\nword_tokens = word_tokenize(text.lower())\n\n
Noktalama i\u015faretlerini ve say\u0131lar\u0131 temizleme<\/h2>\ncleaned_tokens = [re.sub(r'[^a-z\u011f\u00fc\u015f\u00f6\u00e7\u0131\u0130]', '', w) for w in word_tokens] # T\u00fcrk\u00e7e karakterleri de dahil et\ncleaned_tokens = [w for w in cleaned_tokens if w] # Bo\u015f stringleri kald\u0131r\n\nprint(\"\\nTemizlenmi\u015f Belirte\u00e7ler (Noktalama ve Say\u0131 Yok):\")\nprint(cleaned_tokens)\n
\u00c7\u0131kt\u0131:<\/h2>\n
Temizlenmi\u015f Belirte\u00e7ler (Noktalama ve Say\u0131 Yok):<\/h2>\n
['nltk', 'ile', 'y\u0131l\u0131nda', 'nlp', '\u00f6\u011frenmek', '\u00e7ok', 'keyifli', 'bu', 'konuda', 'eminim']<\/code><\/pre>\n<\/h2>\n
K\u00f6k Bulma (Stemming)<\/h4>\n
Stemming, kelimelerin eklerini atarak k\u00f6k formlar\u0131na indirgeme i\u015flemidir. Ama\u00e7, ayn\u0131 k\u00f6kten t\u00fcreyen farkl\u0131 kelimeleri (\u00f6rn. “ko\u015fmak”, “ko\u015fucu”, “ko\u015fuyor”) tek bir formda temsil etmektir. NLTK, Porter Stemmer ve Snowball Stemmer gibi algoritmalar sunar. Snowball Stemmer, \u00e7e\u015fitli diller i\u00e7in destek sa\u011flar.<\/p>\n
from nltk.stem import PorterStemmer, SnowballStemmer\n\nporter = PorterStemmer()\nsnowball = SnowballStemmer(\"english\") # \u0130ngilizce i\u00e7in\n\nwords_en = [\"running\", \"runner\", \"runs\", \"easily\", \"fairly\"]\nprint(\"\\nPorter Stemmer (\u0130ngilizce):\")\nfor w in words_en:\n    print(f\"{w} -> {porter.stem(w)}\")\n
\u00c7\u0131kt\u0131:<\/h2>\n
Porter Stemmer (\u0130ngilizce):<\/h2>\n
running -> run<\/h2>\n
runner -> runner<\/h2>\n
runs -> run<\/h2>\n
easily -> easili<\/h2>\n
fairly -> fairli<\/h2>\n\nprint(\"\\nSnowball Stemmer (\u0130ngilizce):\")\nfor w in words_en:\n    print(f\"{w} -> {snowball.stem(w)}\")\n
\u00c7\u0131kt\u0131:<\/h2>\n
Snowball Stemmer (\u0130ngilizce):<\/h2>\n
running -> run<\/h2>\n
runner -> runner<\/h2>\n
runs -> run<\/h2>\n
easily -> easili<\/h2>\n
fairly -> fairli<\/code><\/pre>\n<\/h2>\n
NLTK’n\u0131n do\u011frudan T\u00fcrk\u00e7e i\u00e7in g\u00fc\u00e7l\u00fc bir Stemmer’\u0131 yoktur. T\u00fcrk\u00e7e, sondan eklemeli bir dil oldu\u011fu i\u00e7in stemming i\u015flemleri daha karma\u015f\u0131kt\u0131r ve \u00f6zel olarak T\u00fcrk\u00e7e i\u00e7in geli\u015ftirilmi\u015f ara\u00e7lar (\u00f6rn. Zemberek) daha iyi sonu\u00e7lar verir.<\/p>\n
Lemmatizasyon (Lemmatization)<\/h4>\n
Lemmatizasyon da kelimeleri temel formlar\u0131na indirgeme i\u015flemidir, ancak stemming’den farkl\u0131 olarak kelimenin s\u00f6zl\u00fckteki ger\u00e7ek k\u00f6k formunu (lemma) bulmaya \u00e7al\u0131\u015f\u0131r. Bu, kelimenin morfolojik analizini gerektirir ve genellikle daha do\u011fru sonu\u00e7lar verir. \u00d6rne\u011fin, “better” kelimesinin lemmas\u0131 “good” iken, stemming sadece “bet” sonucunu verebilir. Lemmatizasyon i\u00e7in genellikle kelimenin Part-of-Speech (POS) etiketi de kullan\u0131l\u0131r.<\/p>\n
from nltk.stem import WordNetLemmatizer\nfrom nltk.corpus import wordnet\n\nlemmatizer = WordNetLemmatizer()\n\nwords_en = [\"cats\", \"cacti\", \"geese\", \"rocks\", \"better\", \"running\"]\n\nprint(\"\\nWordNet Lemmatizer (POS etiketi olmadan):\")\nfor w in words_en:\n    print(f\"{w} -> {lemmatizer.lemmatize(w)}\")\n
\u00c7\u0131kt\u0131:<\/h2>\n
WordNet Lemmatizer (POS etiketi olmadan):<\/h2>\n
cats -> cat<\/h2>\n
cacti -> cactus<\/h2>\n
geese -> goose<\/h2>\n
rocks -> rock<\/h2>\n
better -> better<\/h2>\n
running -> running<\/h2>\n\n
POS etiketleriyle lemmatizasyon (daha do\u011fru sonu\u00e7lar i\u00e7in)<\/h2>\nprint(\"\\nWordNet Lemmatizer (POS etiketi ile):\")\nprint(f\"better (adj) -> {lemmatizer.lemmatize('better', pos=wordnet.ADJ)}\")\nprint(f\"running (verb) -> {lemmatizer.lemmatize('running', pos=wordnet.VERB)}\")\nprint(f\"running (noun) -> {lemmatizer.lemmatize('running', pos=wordnet.NOUN)}\")\n
\u00c7\u0131kt\u0131:<\/h2>\n
WordNet Lemmatizer (POS etiketi ile):<\/h2>\n
better (adj) -> good<\/h2>\n
running (verb) -> run<\/h2>\n
running (noun) -> running<\/code><\/pre>\n<\/h2>\n
Stemming gibi, NLTK’n\u0131n WordNetLemmatizer’\u0131 da \u0130ngilizce i\u00e7in tasarlanm\u0131\u015ft\u0131r. T\u00fcrk\u00e7e i\u00e7in NLTK’da do\u011frudan etkili bir lemmatizer bulunmamaktad\u0131r.<\/p>\n
Metin Analizi ve \u00d6zellik \u00c7\u0131karma<\/h3>\n
Metin \u00f6n i\u015fleme ad\u0131mlar\u0131n\u0131n ard\u0131ndan, NLTK metinler \u00fczerinde \u00e7e\u015fitli analizler yapma ve anlaml\u0131 \u00f6zellikler \u00e7\u0131karma imkan\u0131 sunar.<\/p>\n
POS Etiketleme (Part-of-Speech Tagging)<\/h4>\n
POS etiketleme, bir metindeki her kelimeye dilbilgisel kategorisini (isim, fiil, s\u0131fat, zarf vb.) atama i\u015flemidir. Bu etiketler, kelimelerin c\u00fcmledeki rol\u00fcn\u00fc anlamak ve daha ileri analizler yapmak i\u00e7in kritik \u00f6neme sahiptir.<\/p>\n
from nltk.tokenize import word_tokenize\nfrom nltk import pos_tag\n\ntext_en = \"The quick brown fox jumps over the lazy dog.\"\ntokens_en = word_tokenize(text_en)\npos_tags_en = pos_tag(tokens_en)\n\nprint(\"\\nPOS Etiketleme (\u0130ngilizce):\")\nprint(pos_tags_en)\n
\u00c7\u0131kt\u0131:<\/h2>\n
POS Etiketleme (\u0130ngilizce):<\/h2>\n
[('The', 'DT'), ('quick', 'JJ'), ('brown', 'JJ'), ('fox', 'NN'), ('jumps', 'VBZ'), ('over', 'IN'), ('the', 'DT'), ('lazy', 'JJ'), ('dog', 'NN'), ('.', '.')]<\/code><\/pre>\n<\/h2>\n
NLTK’n\u0131n pos_tag<\/code> fonksiyonu, Penn Treebank etiket setini kullan\u0131r ve \u0130ngilizce metinler i\u00e7in olduk\u00e7a etkilidir. T\u00fcrk\u00e7e gibi farkl\u0131 dil yap\u0131lar\u0131na sahip diller i\u00e7in do\u011frudan bu fonksiyonu kullanmak genellikle do\u011fru sonu\u00e7lar vermez. T\u00fcrk\u00e7e i\u00e7in \u00f6zel olarak e\u011fitilmi\u015f POS etiketleyicileri gereklidir.<\/p>\n
Adland\u0131r\u0131lm\u0131\u015f Varl\u0131k Tan\u0131ma (Named Entity Recognition – NER)<\/h4>\n
NER, bir metindeki adland\u0131r\u0131lm\u0131\u015f varl\u0131klar\u0131 (ki\u015fi adlar\u0131, yer adlar\u0131, kurulu\u015flar, tarihler vb.) tan\u0131mlama ve s\u0131n\u0131fland\u0131rma i\u015flemidir.<\/p>\n
from nltk.tokenize import word_tokenize\nfrom nltk import pos_tag, ne_chunk\n\ntext_en = \"Barack Obama was born in Hawaii and served as the 44th President of the United States.\"\ntokens_en = word_tokenize(text_en)\npos_tags_en = pos_tag(tokens_en)\nnamed_entities = ne_chunk(pos_tags_en)\n\nprint(\"\\nAdland\u0131r\u0131lm\u0131\u015f Varl\u0131k Tan\u0131ma (\u0130ngilizce):\")\nprint(named_entities)\n
\u00c7\u0131kt\u0131 (a\u011fa\u00e7 yap\u0131s\u0131):<\/h2>\n
(S<\/h2>\n
(PERSON Barack Obama)<\/h2>\n
was<\/h2>\n
born<\/h2>\n
in<\/h2>\n
(GPE Hawaii)<\/h2>\n
and<\/h2>\n
served<\/h2>\n
as<\/h2>\n
the<\/h2>\n
44th<\/h2>\n
President<\/h2>\n
of<\/h2>\n
the<\/h2>\n
(GPE United States)<\/h2>\n
.)<\/code><\/pre>\n<\/h2>\n
\u00c7\u0131kt\u0131, bir a\u011fa\u00e7 yap\u0131s\u0131 olarak gelir ve varl\u0131klar\u0131 (PERSON, GPE – Geopolitical Entity) g\u00f6sterir. NLTK’n\u0131n NER modeli de \u0130ngilizce odakl\u0131d\u0131r ve T\u00fcrk\u00e7e i\u00e7in \u00f6zel bir model gereklidir.<\/p>\n
Frekans Da\u011f\u0131l\u0131m\u0131 (Frequency Distribution)<\/h4>\n
Bir metindeki kelimelerin veya belirte\u00e7lerin ne s\u0131kl\u0131kla ge\u00e7ti\u011fini analiz etmek, metnin ana temalar\u0131n\u0131 anlamak i\u00e7in \u00e7ok faydal\u0131d\u0131r. NLTK’n\u0131n FreqDist<\/code> s\u0131n\u0131f\u0131 bu g\u00f6revi kolayla\u015ft\u0131r\u0131r.<\/p>\n
from nltk.probability import FreqDist\nfrom nltk.tokenize import word_tokenize\nfrom nltk.corpus import stopwords\nimport matplotlib.pyplot as plt\n\ntext_tr = \"Do\u011fal dil i\u015fleme \u00f6\u011frenmek \u00e7ok keyifli. NLTK ile do\u011fal dil i\u015fleme \u00f6\u011frenmek daha da keyifli. Python do\u011fal dil i\u015fleme i\u00e7in harika bir dil.\"\ntokens_tr = word_tokenize(text_tr.lower())\n\n
T\u00fcrk\u00e7e durma kelimeleri (\u00f6rnek)<\/h2>\nturkish_stopwords = [\n    \"bir\", \"bu\", \"\u00e7ok\", \"i\u00e7in\", \"ile\", \"de\", \"da\", \"ve\", \"daha\", \"\u00f6\u011frenmek\"\n]\nfiltered_tokens_tr = [w for w in tokens_tr if w.isalpha() and w not in turkish_stopwords]\n\nfdist = FreqDist(filtered_tokens_tr)\n\nprint(\"\\nEn S\u0131k Ge\u00e7en Kelimeler (T\u00fcrk\u00e7e):\")\nprint(fdist.most_common(5))\n
\u00c7\u0131kt\u0131:<\/h2>\n
En S\u0131k Ge\u00e7en Kelimeler (T\u00fcrk\u00e7e):<\/h2>\n
[('do\u011fal', 3), ('dil', 3), ('i\u015fleme', 3), ('nltk', 1), ('keyifli', 2)]<\/h2>\n\n
Frekans da\u011f\u0131l\u0131m\u0131n\u0131 g\u00f6rselle\u015ftirme<\/h2>\n
fdist.plot(10, cumulative=False, title=\"En S\u0131k Ge\u00e7en Kelimeler\")<\/h2>\n
plt.show()<\/code><\/pre>\n<\/h2>\n
Yukar\u0131daki kod, matplotlib<\/code> k\u00fct\u00fcphanesini kullanarak en s\u0131k ge\u00e7en kelimelerin bir grafi\u011fini \u00e7izebilir.<\/p>\n
Konkordans (Concordance)<\/h4>\n
Konkordans, belirli bir kelimenin metin i\u00e7inde hangi ba\u011flamlarda kullan\u0131ld\u0131\u011f\u0131n\u0131 g\u00f6sterir. Bu, bir kelimenin farkl\u0131 anlamlar\u0131n\u0131 veya kullan\u0131m \u015fekillerini incelemek i\u00e7in kullan\u0131\u015fl\u0131d\u0131r.<\/p>\n
from nltk.text import Text\nfrom nltk.tokenize import word_tokenize\n\nlong_text = \"NLTK, do\u011fal dil i\u015fleme \u00f6\u011frenmek i\u00e7in harika bir ba\u015flang\u0131\u00e7 noktas\u0131d\u0131r. Python ile NLTK kullanarak metin analizi yapabilir, kelimelerin frekans\u0131n\u0131 \u00e7\u0131karabilir ve dilbilgisel yap\u0131lar\u0131 inceleyebilirsiniz. Do\u011fal dil i\u015fleme d\u00fcnyas\u0131na ad\u0131m atmak i\u00e7in NLTK iyi bir ara\u00e7t\u0131r.\"\ntokens = word_tokenize(long_text.lower())\nnltk_text = Text(tokens)\n\nprint(\"\\n'NLTK' kelimesinin konkordans\u0131:\")\nnltk_text.concordance(\"nltk\")\n
\u00c7\u0131kt\u0131:<\/h2>\n
'NLTK' kelimesinin konkordans\u0131:<\/h2>\n
Displaying 3 of 3 matches:<\/h2>\n
nltk , do\u011fal dil i\u015fleme \u00f6\u011frenmek i\u00e7in harika bir<\/h2>\n
python ile nltk kullanarak metin analizi yapabilir<\/h2>\n
dil i\u015fleme d\u00fcnyas\u0131na ad\u0131m atmak i\u00e7in nltk iyi bir ara\u00e7t\u0131r .<\/code><\/pre>\n<\/h2>\n
Benzer Kelimeler ve Ortak Ba\u011flamlar<\/h4>\n
NLTK Text<\/code> nesnesi, bir kelimeye benzer ba\u011flamlarda kullan\u0131lan di\u011fer kelimeleri (similar<\/code>) veya iki kelimenin ortak ba\u011flamlar\u0131n\u0131 (common_contexts<\/code>) bulmak i\u00e7in de kullan\u0131labilir.<\/p>\n
print(\"\\n'NLTK' kelimesine benzer ba\u011flamlar:\")\nnltk_text.similar(\"nltk\")\n
\u00c7\u0131kt\u0131:<\/h2>\n
'NLTK' kelimesine benzer ba\u011flamlar:<\/h2>\n
dil<\/h2>\n
If you don't see anything, try different words.<\/h2>\n\nprint(\"\\n'NLTK' ve 'dil' kelimelerinin ortak ba\u011flamlar\u0131:\")\nnltk_text.common_contexts([\"nltk\", \"dil\"])\n
\u00c7\u0131kt\u0131:<\/h2>\n
'NLTK' ve 'dil' kelimelerinin ortak ba\u011flamlar\u0131:<\/h2>\n
do\u011fal _ i\u015fleme<\/code><\/pre>\n<\/h2>\n
Bu fonksiyonlar, metnin anlamsal yap\u0131s\u0131n\u0131 ke\u015ffetmek i\u00e7in faydal\u0131 olabilir.<\/p>\n
Dil Modelleri ve S\u0131n\u0131fland\u0131rma<\/h3>\n
NLTK, temel dil modelleri olu\u015fturma ve metin s\u0131n\u0131fland\u0131rma gibi daha ileri NLP g\u00f6revleri i\u00e7in de mod\u00fcller sunar.<\/p>\n
N-gram Modelleri<\/h4>\n
N-gramlar, bir metindeki ard\u0131\u015f\u0131k kelime dizileridir. Genellikle, bir kelimenin kendisinden \u00f6nceki N-1 kelimeye g\u00f6re tahmin edilme olas\u0131l\u0131\u011f\u0131n\u0131 hesaplamak i\u00e7in kullan\u0131l\u0131rlar.<\/p>\n
from nltk.util import ngrams\nfrom nltk.tokenize import word_tokenize\n\ntext = \"Do\u011fal dil i\u015fleme \u00f6\u011frenmek \u00e7ok keyifli.\"\ntokens = word_tokenize(text.lower())\n\n
Bigramlar (2-gram)<\/h2>\nbigrams = list(ngrams(tokens, 2))\nprint(\"\\nBigramlar:\")\nprint(bigrams)\n
\u00c7\u0131kt\u0131:<\/h2>\n
Bigramlar:<\/h2>\n
[('do\u011fal', 'dil'), ('dil', 'i\u015fleme'), ('i\u015fleme', '\u00f6\u011frenmek'), ('\u00f6\u011frenmek', '\u00e7ok'), ('\u00e7ok', 'keyifli'), ('keyifli', '.')]<\/h2>\n\n
Trigramlar (3-gram)<\/h2>\ntrigrams = list(ngrams(tokens, 3))\nprint(\"\\nTrigramlar:\")\nprint(trigrams)\n
\u00c7\u0131kt\u0131:<\/h2>\n
Trigramlar:<\/h2>\n
[('do\u011fal', 'dil', 'i\u015fleme'), ('dil', 'i\u015fleme', '\u00f6\u011frenmek'), ('i\u015fleme', '\u00f6\u011frenmek', '\u00e7ok'), ('\u00f6\u011frenmek', '\u00e7ok', 'keyifli'), ('\u00e7ok', 'keyifli', '.')]<\/code><\/pre>\n<\/h2>\n
N-gramlar, metin \u00fcretimi, konu\u015fma tan\u0131ma ve makine \u00e7evirisi gibi bir\u00e7ok NLP uygulamas\u0131n\u0131n temelini olu\u015fturur.<\/p>\n
Metin S\u0131n\u0131fland\u0131rma<\/h4>\n
NLTK, \u00e7e\u015fitli s\u0131n\u0131fland\u0131rma algoritmalar\u0131 (\u00f6rn. Naive Bayes) i\u00e7in temel bir \u00e7er\u00e7eve sunar. A\u015fa\u011f\u0131daki \u00f6rnek, basit bir metin s\u0131n\u0131fland\u0131r\u0131c\u0131s\u0131n\u0131n nas\u0131l olu\u015fturulaca\u011f\u0131n\u0131 g\u00f6sterir.<\/p>\n
import random\nfrom nltk.corpus import movie_reviews\nfrom nltk.tokenize import word_tokenize\nfrom nltk.classify import NaiveBayesClassifier\nfrom nltk.classify.util import accuracy\n\n
Film yorumlar\u0131 veri setini y\u00fckle (pozitif ve negatif)<\/h2>\n
nltk.download('movie_reviews') # E\u011fer indirilmediyse<\/h2>\n\ndocuments = [(list(movie_reviews.words(fileid)), category)\n             for category in movie_reviews.categories()\n             for fileid in movie_reviews.fileids(category)]\n\nrandom.shuffle(documents)\n\n
T\u00fcm kelimeleri topla ve frekans da\u011f\u0131l\u0131m\u0131n\u0131 olu\u015ftur<\/h2>\nall_words = FreqDist(w.lower() for w in movie_reviews.words())\nword_features = list(all_words.keys())[:2000] # En s\u0131k ge\u00e7en 2000 kelimeyi \u00f6zellik olarak al\n\n
Belge i\u00e7in \u00f6zellik \u00e7\u0131kar\u0131c\u0131 fonksiyon<\/h2>\ndef find_features(document):\n    words = word_tokenize(document) # document zaten kelime listesi, bu ad\u0131m gereksiz olabilir\n    features = {}\n    for w in word_features:\n        features[w] = (w in words) # Kelimenin belgede olup olmad\u0131\u011f\u0131n\u0131 kontrol et\n    return features\n\n
Veri setini \u00f6zellik setlerine d\u00f6n\u00fc\u015ft\u00fcr<\/h2>\nfeaturesets = [(find_features(doc), category) for (doc, category) in documents]\n\n
E\u011fitim ve test setlerini ay\u0131r<\/h2>\ntrain_set, test_set = featuresets[100:], featuresets[:100]\n\n
Naive Bayes s\u0131n\u0131fland\u0131r\u0131c\u0131y\u0131 e\u011fit<\/h2>\nclassifier = NaiveBayesClassifier.train(train_set)\n\nprint(\"\\nNaive Bayes S\u0131n\u0131fland\u0131r\u0131c\u0131 Do\u011frulu\u011fu:\")\nprint(f\"Do\u011fruluk: {accuracy(classifier, test_set) * 100:.2f}%\")\n\nprint(\"\\nEn Bilgilendirici \u00d6zellikler:\")\nclassifier.show_most_informative_features(5)\n
\u00c7\u0131kt\u0131 (\u00f6rnek):<\/h2>\n
Naive Bayes S\u0131n\u0131fland\u0131r\u0131c\u0131 Do\u011frulu\u011fu:<\/h2>\n
Do\u011fruluk: 78.00%<\/h2>\n
# En Bilgilendirici \u00d6zellikler:<\/h2>\n
Most Informative Features<\/h2>\n
outstanding = True              pos : neg    =     10.6 : 1.0<\/h2>\n
lame = True              neg : pos    =      9.8 : 1.0<\/h2>\n
magnificent = True              pos : neg    =      8.6 : 1.0<\/h2>\n
seagal = True              neg : pos    =      8.4 : 1.0<\/h2>\n
poor = True              neg : pos    =      7.6 : 1.0<\/code><\/pre>\n<\/h2>\n
Bu \u00f6rnek, NLTK ile temel bir metin s\u0131n\u0131fland\u0131rma pipeline’\u0131n\u0131n nas\u0131l olu\u015fturulaca\u011f\u0131n\u0131 g\u00f6sterir. Daha b\u00fcy\u00fck ve karma\u015f\u0131k veri setleri i\u00e7in scikit-learn<\/code> veya Keras<\/code>\/PyTorch<\/code> gibi k\u00fct\u00fcphanelerle entegrasyon daha yayg\u0131n ve performansl\u0131d\u0131r.<\/p>\n
NLTK’n\u0131n S\u0131n\u0131rl\u0131l\u0131klar\u0131 ve Alternatifler<\/h3>\n
NLTK, do\u011fal dil i\u015flemenin temellerini \u00f6\u011frenmek ve k\u00fc\u00e7\u00fck \u00f6l\u00e7ekli projeler i\u00e7in harika bir ara\u00e7 olsa da, baz\u0131 s\u0131n\u0131rl\u0131l\u0131klar\u0131 vard\u0131r:<\/p>\n
*   Performans:<\/strong> B\u00fcy\u00fck veri setleri ve \u00fcretim ortamlar\u0131 i\u00e7in NLTK’n\u0131n performans\u0131 yetersiz kalabilir. \u00d6zellikle belirte\u00e7lere ay\u0131rma ve POS etiketleme gibi i\u015flemler daha yava\u015f \u00e7al\u0131\u015fabilir.
\n*   Dil Deste\u011fi:<\/strong> NLTK, a\u011f\u0131rl\u0131kl\u0131 olarak \u0130ngilizce odakl\u0131d\u0131r. T\u00fcrk\u00e7e gibi farkl\u0131 dil yap\u0131lar\u0131na sahip diller i\u00e7in do\u011frudan g\u00fc\u00e7l\u00fc ve haz\u0131r modelleri bulunmamaktad\u0131r. Bu t\u00fcr diller i\u00e7in genellikle \u00f6zel olarak e\u011fitilmi\u015f modeller veya ba\u015fka k\u00fct\u00fcphaneler kullan\u0131l\u0131r.
\n*   Modern NLP Yakla\u015f\u0131mlar\u0131:<\/strong> Derin \u00f6\u011frenme tabanl\u0131 NLP modelleri (\u00f6rn. Transformer modelleri) NLTK’n\u0131n kapsam\u0131 d\u0131\u015f\u0131ndad\u0131r. NLTK, istatistiksel ve kural tabanl\u0131 yakla\u015f\u0131mlara daha yatk\u0131nd\u0131r.<\/p>\n
Bu s\u0131n\u0131rl\u0131l\u0131klar g\u00f6z \u00f6n\u00fcne al\u0131nd\u0131\u011f\u0131nda, daha geli\u015fmi\u015f NLP g\u00f6revleri veya \u00fcretim ortamlar\u0131 i\u00e7in a\u015fa\u011f\u0131daki alternatifler de\u011ferlendirilebilir:<\/p>\n
*   spaCy:<\/strong> H\u0131zl\u0131, \u00fcretim ortamlar\u0131 i\u00e7in optimize edilmi\u015f, \u00f6nceden e\u011fitilmi\u015f modellerle gelen modern bir NLP k\u00fct\u00fcphanesidir. \u00d6zellikle \u0130ngilizce ve baz\u0131 di\u011fer diller i\u00e7in g\u00fc\u00e7l\u00fc dil modellerine sahiptir.
\n*   Gensim:<\/strong> Konu modelleme (Latent Dirichlet Allocation – LDA) ve kelime vekt\u00f6rleri (Word2Vec, Doc2Vec) gibi istatistiksel semantik modelleme g\u00f6revleri i\u00e7in optimize edilmi\u015ftir.
\n*   Hugging Face Transformers:<\/strong> Derin \u00f6\u011frenme tabanl\u0131, son teknoloji Transformer modellerini (BERT, GPT, T5 vb.) kullanmak i\u00e7in pop\u00fcler bir k\u00fct\u00fcphanedir. Farkl\u0131 diller i\u00e7in geni\u015f bir model yelpazesi sunar.
\n*   Zemberek (Python i\u00e7in pyezemberek):<\/strong> T\u00fcrk\u00e7e do\u011fal dil i\u015fleme i\u00e7in \u00f6zel olarak geli\u015ftirilmi\u015f g\u00fc\u00e7l\u00fc bir k\u00fct\u00fcphanedir. K\u00f6k bulma, lemmatizasyon, morfolojik analiz ve heceleme gibi T\u00fcrk\u00e7e’ye \u00f6zg\u00fc g\u00f6revlerde NLTK’dan \u00e7ok daha ba\u015far\u0131l\u0131d\u0131r.<\/p>\n
NLTK, bu k\u00fct\u00fcphanelerin temelini anlamak ve NLP’nin nas\u0131l \u00e7al\u0131\u015ft\u0131\u011f\u0131na dair sa\u011flam bir temel olu\u015fturmak i\u00e7in m\u00fckemmel bir ba\u015flang\u0131\u00e7 noktas\u0131d\u0131r. Genellikle, bir projeye NLTK ile ba\u015flan\u0131r ve ihtiya\u00e7lar do\u011frultusunda daha spesifik veya performans odakl\u0131 k\u00fct\u00fcphanelere ge\u00e7i\u015f yap\u0131l\u0131r.<\/p>\n
Sonu\u00e7<\/h3>\n
Natural Language Toolkit (NLTK), Python programlama dili ile do\u011fal dil i\u015fleme d\u00fcnyas\u0131na ad\u0131m atmak isteyen herkes i\u00e7in g\u00fc\u00e7l\u00fc ve eri\u015filebilir bir kap\u0131d\u0131r. Bu makale boyunca, NLTK’n\u0131n kurulumundan ba\u015flayarak, metin \u00f6n i\u015fleme (belirte\u00e7lere ay\u0131rma, durma kelimeleri, k\u00f6k bulma, lemmatizasyon), temel metin analizi (POS etiketleme, NER, frekans da\u011f\u0131l\u0131m\u0131) ve hatta basit dil modelleri ile s\u0131n\u0131fland\u0131rma gibi bir\u00e7ok temel NLP g\u00f6revini nas\u0131l yerine getirebilece\u011finizi \u00f6\u011frendik.<\/p>\n
NLTK, \u00f6zellikle e\u011fitim ama\u00e7l\u0131, prototipleme ve temel ara\u015ft\u0131rmalar i\u00e7in paha bi\u00e7ilmez bir ara\u00e7t\u0131r. Zengin veri setleri ve algoritmalar\u0131 sayesinde, do\u011fal dil verilerini anlamak ve i\u015flemek i\u00e7in gerekli temel yetkinlikleri kazanman\u0131z\u0131 sa\u011flar. Ancak, b\u00fcy\u00fck \u00f6l\u00e7ekli uygulamalar, y\u00fcksek performans gerektiren sistemler veya T\u00fcrk\u00e7e gibi \u0130ngilizce d\u0131\u015f\u0131ndaki diller i\u00e7in daha \u00f6zelle\u015fmi\u015f ve modern k\u00fct\u00fcphanelerin (spaCy, Gensim, Hugging Face Transformers, Zemberek) de var oldu\u011funu unutmamak \u00f6nemlidir.<\/p>\n
NLP alan\u0131 s\u00fcrekli geli\u015fmekte ve yeni teknolojilerle zenginle\u015fmektedir. NLTK ile edindi\u011finiz bu temel bilgi birikimi, sizi daha karma\u015f\u0131k ve heyecan verici NLP projelerine haz\u0131rlayacak sa\u011flam bir temel olu\u015fturacakt\u0131r. Dilin derinliklerini ke\u015ffetmeye ve metin verilerinden anlaml\u0131 i\u00e7g\u00f6r\u00fcler \u00e7\u0131karmaya devam etmek i\u00e7in NLTK, yolculu\u011funuzda \u00f6nemli bir ilk ad\u0131md\u0131r.<\/p>\n
<\/body><\/p>\n","protected":false},"excerpt":{"rendered":"Python 3’te Dil Verileriyle \u00c7al\u0131\u015fmak: Do\u011fal Dil \u0130\u015fleme \u0130\u00e7in NLTK K\u00fct\u00fcphanesi\nDo\u011fal Dil \u0130\u015fleme (NLP), bilgisayarlar\u0131n insan dilini anla","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"csco_page_header_type":"","csco_page_load_nextpost":"","csco_page_subscribe_form":"","csco_page_contact_form":"","footnotes":""},"categories":[1403],"tags":[],"class_list":{"0":"post-34968","1":"post","2":"type-post","3":"status-publish","4":"format-standard","6":"category-python","7":"cs-entry","8":"cs-video-wrap"},"yoast_head":"\nPython 3'te Dil Verileriyle \u00c7al\u0131\u015fmak: Do\u011fal Dil \u0130\u015fleme \u0130\u00e7in NLTK K\u00fct\u00fcphanesi - Kodlar\u0131n Gizemli D\u00fcnyas\u0131<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/fatihsoysal.com\/blog\/python-3te-dil-verileriyle-calismak-dogal-dil-isleme-icin-nltk-kutuphanesi\/\" \/>\n<meta property=\"og:locale\" content=\"tr_TR\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Python 3'te Dil Verileriyle \u00c7al\u0131\u015fmak: Do\u011fal Dil \u0130\u015fleme \u0130\u00e7in NLTK K\u00fct\u00fcphanesi\" \/>\n<meta property=\"og:description\" content=\"Python 3'te Dil Verileriyle \u00c7al\u0131\u015fmak: Do\u011fal Dil \u0130\u015fleme \u0130\u00e7in NLTK K\u00fct\u00fcphanesi Do\u011fal Dil \u0130\u015fleme (NLP), bilgisayarlar\u0131n insan dilini anla\" \/>\n<meta property=\"og:url\" content=\"https:\/\/fatihsoysal.com\/blog\/python-3te-dil-verileriyle-calismak-dogal-dil-isleme-icin-nltk-kutuphanesi\/\" \/>\n<meta property=\"og:site_name\" content=\"Kodlar\u0131n Gizemli D\u00fcnyas\u0131\" \/>\n<meta property=\"article:published_time\" content=\"2025-11-24T04:40:46+00:00\" \/>\n<meta name=\"author\" content=\"Fatih Soysal\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Yazan:\" \/>\n\t<meta name=\"twitter:data1\" content=\"Fatih Soysal\" \/>\n\t<meta name=\"twitter:label2\" content=\"Tahmini okuma s\u00fcresi\" \/>\n\t<meta name=\"twitter:data2\" content=\"19 dakika\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\/\/fatihsoysal.com\/blog\/python-3te-dil-verileriyle-calismak-dogal-dil-isleme-icin-nltk-kutuphanesi\/#article\",\"isPartOf\":{\"@id\":\"https:\/\/fatihsoysal.com\/blog\/python-3te-dil-verileriyle-calismak-dogal-dil-isleme-icin-nltk-kutuphanesi\/\"},\"author\":{\"name\":\"Fatih Soysal\",\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/002a254750921dcfd568a99e48240dd1\"},\"headline\":\"Python 3’te Dil Verileriyle \u00c7al\u0131\u015fmak: Do\u011fal Dil \u0130\u015fleme \u0130\u00e7in NLTK K\u00fct\u00fcphanesi\",\"datePublished\":\"2025-11-24T04:40:46+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\/\/fatihsoysal.com\/blog\/python-3te-dil-verileriyle-calismak-dogal-dil-isleme-icin-nltk-kutuphanesi\/\"},\"wordCount\":2149,\"commentCount\":0,\"publisher\":{\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/002a254750921dcfd568a99e48240dd1\"},\"articleSection\":[\"Python\"],\"inLanguage\":\"tr\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\/\/fatihsoysal.com\/blog\/python-3te-dil-verileriyle-calismak-dogal-dil-isleme-icin-nltk-kutuphanesi\/#respond\"]}],\"copyrightYear\":\"2025\",\"copyrightHolder\":{\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#organization\"}},{\"@type\":\"WebPage\",\"@id\":\"https:\/\/fatihsoysal.com\/blog\/python-3te-dil-verileriyle-calismak-dogal-dil-isleme-icin-nltk-kutuphanesi\/\",\"url\":\"https:\/\/fatihsoysal.com\/blog\/python-3te-dil-verileriyle-calismak-dogal-dil-isleme-icin-nltk-kutuphanesi\/\",\"name\":\"Python 3'te Dil Verileriyle \u00c7al\u0131\u015fmak: Do\u011fal Dil \u0130\u015fleme \u0130\u00e7in NLTK K\u00fct\u00fcphanesi - Kodlar\u0131n Gizemli D\u00fcnyas\u0131\",\"isPartOf\":{\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#website\"},\"datePublished\":\"2025-11-24T04:40:46+00:00\",\"breadcrumb\":{\"@id\":\"https:\/\/fatihsoysal.com\/blog\/python-3te-dil-verileriyle-calismak-dogal-dil-isleme-icin-nltk-kutuphanesi\/#breadcrumb\"},\"inLanguage\":\"tr\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/fatihsoysal.com\/blog\/python-3te-dil-verileriyle-calismak-dogal-dil-isleme-icin-nltk-kutuphanesi\/\"]}]},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/fatihsoysal.com\/blog\/python-3te-dil-verileriyle-calismak-dogal-dil-isleme-icin-nltk-kutuphanesi\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Anasayfa\",\"item\":\"https:\/\/fatihsoysal.com\/blog\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Python 3’te Dil Verileriyle \u00c7al\u0131\u015fmak: Do\u011fal Dil \u0130\u015fleme \u0130\u00e7in NLTK K\u00fct\u00fcphanesi\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#website\",\"url\":\"https:\/\/fatihsoysal.com\/blog\/\",\"name\":\"Fatihsoysal.com\",\"description\":\"Blog - Yaz\u0131l\u0131m D\u00fcnyas\u0131 Tecr\u00fcbelerim\",\"publisher\":{\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/002a254750921dcfd568a99e48240dd1\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/fatihsoysal.com\/blog\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"tr\"},{\"@type\":[\"Person\",\"Organization\"],\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/002a254750921dcfd568a99e48240dd1\",\"name\":\"Fatih Soysal\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"tr\",\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/fatihsoysal.com\/blog\/wp-content\/uploads\/2024\/04\/cropped-replicate-prediction-3kgg1hgjn5rgp0cf0p5tr0jw7w-1.png\",\"contentUrl\":\"https:\/\/fatihsoysal.com\/blog\/wp-content\/uploads\/2024\/04\/cropped-replicate-prediction-3kgg1hgjn5rgp0cf0p5tr0jw7w-1.png\",\"width\":512,\"height\":512,\"caption\":\"Fatih Soysal\"},\"logo\":{\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/image\/\"},\"description\":\"Kullan\u0131m ve kodlama m\u00fckemmeliyetini odak alan uygulamalar olu\u015fturma deneyimine sahip, profesyonel olarak 15+ y\u0131l \u00fczeri deneyime sahip bir yaz\u0131l\u0131m m\u00fchendisi.\",\"url\":\"https:\/\/fatihsoysal.com\/blog\/author\/fatihsoysal\/\"}]}<\/script>\n<!-- \/ Yoast SEO Premium plugin. -->","yoast_head_json":{"title":"Python 3'te Dil Verileriyle \u00c7al\u0131\u015fmak: Do\u011fal Dil \u0130\u015fleme \u0130\u00e7in NLTK K\u00fct\u00fcphanesi - Kodlar\u0131n Gizemli D\u00fcnyas\u0131","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/fatihsoysal.com\/blog\/python-3te-dil-verileriyle-calismak-dogal-dil-isleme-icin-nltk-kutuphanesi\/","og_locale":"tr_TR","og_type":"article","og_title":"Python 3'te Dil Verileriyle \u00c7al\u0131\u015fmak: Do\u011fal Dil \u0130\u015fleme \u0130\u00e7in NLTK K\u00fct\u00fcphanesi","og_description":"Python 3'te Dil Verileriyle \u00c7al\u0131\u015fmak: Do\u011fal Dil \u0130\u015fleme \u0130\u00e7in NLTK K\u00fct\u00fcphanesi Do\u011fal Dil \u0130\u015fleme (NLP), bilgisayarlar\u0131n insan dilini anla","og_url":"https:\/\/fatihsoysal.com\/blog\/python-3te-dil-verileriyle-calismak-dogal-dil-isleme-icin-nltk-kutuphanesi\/","og_site_name":"Kodlar\u0131n Gizemli D\u00fcnyas\u0131","article_published_time":"2025-11-24T04:40:46+00:00","author":"Fatih Soysal","twitter_card":"summary_large_image","twitter_misc":{"Yazan:":"Fatih Soysal","Tahmini okuma s\u00fcresi":"19 dakika"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/fatihsoysal.com\/blog\/python-3te-dil-verileriyle-calismak-dogal-dil-isleme-icin-nltk-kutuphanesi\/#article","isPartOf":{"@id":"https:\/\/fatihsoysal.com\/blog\/python-3te-dil-verileriyle-calismak-dogal-dil-isleme-icin-nltk-kutuphanesi\/"},"author":{"name":"Fatih Soysal","@id":"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/002a254750921dcfd568a99e48240dd1"},"headline":"Python 3’te Dil Verileriyle \u00c7al\u0131\u015fmak: Do\u011fal Dil \u0130\u015fleme \u0130\u00e7in NLTK K\u00fct\u00fcphanesi","datePublished":"2025-11-24T04:40:46+00:00","mainEntityOfPage":{"@id":"https:\/\/fatihsoysal.com\/blog\/python-3te-dil-verileriyle-calismak-dogal-dil-isleme-icin-nltk-kutuphanesi\/"},"wordCount":2149,"commentCount":0,"publisher":{"@id":"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/002a254750921dcfd568a99e48240dd1"},"articleSection":["Python"],"inLanguage":"tr","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/fatihsoysal.com\/blog\/python-3te-dil-verileriyle-calismak-dogal-dil-isleme-icin-nltk-kutuphanesi\/#respond"]}],"copyrightYear":"2025","copyrightHolder":{"@id":"https:\/\/fatihsoysal.com\/blog\/#organization"}},{"@type":"WebPage","@id":"https:\/\/fatihsoysal.com\/blog\/python-3te-dil-verileriyle-calismak-dogal-dil-isleme-icin-nltk-kutuphanesi\/","url":"https:\/\/fatihsoysal.com\/blog\/python-3te-dil-verileriyle-calismak-dogal-dil-isleme-icin-nltk-kutuphanesi\/","name":"Python 3'te Dil Verileriyle \u00c7al\u0131\u015fmak: Do\u011fal Dil \u0130\u015fleme \u0130\u00e7in NLTK K\u00fct\u00fcphanesi - Kodlar\u0131n Gizemli D\u00fcnyas\u0131","isPartOf":{"@id":"https:\/\/fatihsoysal.com\/blog\/#website"},"datePublished":"2025-11-24T04:40:46+00:00","breadcrumb":{"@id":"https:\/\/fatihsoysal.com\/blog\/python-3te-dil-verileriyle-calismak-dogal-dil-isleme-icin-nltk-kutuphanesi\/#breadcrumb"},"inLanguage":"tr","potentialAction":[{"@type":"ReadAction","target":["https:\/\/fatihsoysal.com\/blog\/python-3te-dil-verileriyle-calismak-dogal-dil-isleme-icin-nltk-kutuphanesi\/"]}]},{"@type":"BreadcrumbList","@id":"https:\/\/fatihsoysal.com\/blog\/python-3te-dil-verileriyle-calismak-dogal-dil-isleme-icin-nltk-kutuphanesi\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Anasayfa","item":"https:\/\/fatihsoysal.com\/blog\/"},{"@type":"ListItem","position":2,"name":"Python 3’te Dil Verileriyle \u00c7al\u0131\u015fmak: Do\u011fal Dil \u0130\u015fleme \u0130\u00e7in NLTK K\u00fct\u00fcphanesi"}]},{"@type":"WebSite","@id":"https:\/\/fatihsoysal.com\/blog\/#website","url":"https:\/\/fatihsoysal.com\/blog\/","name":"Fatihsoysal.com","description":"Blog - Yaz\u0131l\u0131m D\u00fcnyas\u0131 Tecr\u00fcbelerim","publisher":{"@id":"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/002a254750921dcfd568a99e48240dd1"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/fatihsoysal.com\/blog\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"tr"},{"@type":["Person","Organization"],"@id":"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/002a254750921dcfd568a99e48240dd1","name":"Fatih Soysal","image":{"@type":"ImageObject","inLanguage":"tr","@id":"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/image\/","url":"https:\/\/fatihsoysal.com\/blog\/wp-content\/uploads\/2024\/04\/cropped-replicate-prediction-3kgg1hgjn5rgp0cf0p5tr0jw7w-1.png","contentUrl":"https:\/\/fatihsoysal.com\/blog\/wp-content\/uploads\/2024\/04\/cropped-replicate-prediction-3kgg1hgjn5rgp0cf0p5tr0jw7w-1.png","width":512,"height":512,"caption":"Fatih Soysal"},"logo":{"@id":"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/image\/"},"description":"Kullan\u0131m ve kodlama m\u00fckemmeliyetini odak alan uygulamalar olu\u015fturma deneyimine sahip, profesyonel olarak 15+ y\u0131l \u00fczeri deneyime sahip bir yaz\u0131l\u0131m m\u00fchendisi.","url":"https:\/\/fatihsoysal.com\/blog\/author\/fatihsoysal\/"}]}},"yoast_meta":{"yoast_wpseo_title":"","yoast_wpseo_metadesc":"","yoast_wpseo_canonical":""},"amp_enabled":true,"_links":{"self":[{"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/posts\/34968","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/comments?post=34968"}],"version-history":[{"count":1,"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/posts\/34968\/revisions"}],"predecessor-version":[{"id":34969,"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/posts\/34968\/revisions\/34969"}],"wp:attachment":[{"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/media?parent=34968"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/categories?post=34968"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/tags?post=34968"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}