TextBlob

from textblob import TextBlob

text = "Natural language processing is a fascinating field that bridges linguistics and AI."
blob = TextBlob(text)

print(type(blob))       # <class 'textblob.blob.TextBlob'>
print(blob.words)       # WordList(['Natural', 'language', 'processing', ...])
print(blob.sentences)   # [Sentence("Natural language processing...")]

from textblob import TextBlob

reviews = [
    "This new AI model is absolutely amazing and incredibly fast!",
    "The documentation is confusing and the examples are outdated.",
    "The library works as expected. Nothing special.",
    "Terrible performance. I wouldn't recommend it to anyone.",
    "Great library for beginners. The API is clean and intuitive."
]

print(f"{'Review':<55} {'Polarity':>9} {'Subjectivity':>13} {'Sentiment':>10}")
print("-" * 95)

for review in reviews:
    blob = TextBlob(review)
    pol = blob.sentiment.polarity
    sub = blob.sentiment.subjectivity
    sentiment = "Positive" if pol > 0.1 else "Negative" if pol < -0.1 else "Neutral"
    print(f"{review[:54]:<55} {pol:>9.3f} {sub:>13.3f} {sentiment:>10}")

from textblob import TextBlob

blob = TextBlob("The deep learning model transformed our NLP pipeline significantly.")

for word, tag in blob.tags:
    print(f"{word:<20} {tag}")

# The                  DT
# deep                 JJ
# learning             NN
# model                NN
# transformed          VBD
# NLP                  NNP
# pipeline             NN
# significantly        RB

from textblob import TextBlob

text = """
Large language models and retrieval-augmented generation systems
have transformed enterprise NLP applications in recent years.
"""

blob = TextBlob(text)
print("Noun phrases:", blob.noun_phrases)
# WordList(['large language models', 'retrieval-augmented generation systems',
#           'enterprise nlp applications', 'recent years'])

from textblob import TextBlob

misspelled_texts = [
    "Natral languaj procesing is usefull for text analisis.",
    "The modle achieeved excelent resutls on the benchmrk.",
    "Sentimnt anlysis helps buisnesses undrstand custmer feeback."
]

for text in misspelled_texts:
    blob = TextBlob(text)
    corrected = blob.correct()
    print(f"Original:  {text}")
    print(f"Corrected: {corrected}\n")

from textblob import Word

words = ["running", "models", "better", "geese", "phenomena"]

for word_str in words:
    word = Word(word_str)
    print(f"{word_str:<15} singular: {word.singularize():<15} lemma: {word.lemmatize()}")

# running         singular: running         lemma: running
# models          singular: model           lemma: model
# geese           singular: goose           lemma: goose
# phenomena       singular: phenomenon      lemma: phenomenon

from textblob import TextBlob

# Detect language
texts = [
    "Natural language processing is a field of AI.",
    "Le traitement du langage naturel est un domaine fascinant.",
    "El procesamiento del lenguaje natural transforma la IA.",
    "自然言語処理は人工知能の一分野です。"
]

for text in texts:
    blob = TextBlob(text)
    lang = blob.detect_language()
    print(f"{lang}: {text[:50]}")

# Translate French to English
french_blob = TextBlob("Le modèle de langage comprend le texte humain.")
english = french_blob.translate(to='en')
print(english)
# "The language model understands human text."

from textblob import TextBlob

blob = TextBlob("Transformer models process text using attention mechanisms.")

print("Bigrams:", blob.ngrams(n=2))
print("Trigrams:", blob.ngrams(n=3))

# Bigrams: [WordList(['Transformer', 'models']), WordList(['models', 'process']),...]
# Trigrams: [WordList(['Transformer', 'models', 'process']),...]

from textblob.classifiers import NaiveBayesClassifier

train = [
    ("The model performance is excellent!", "positive"),
    ("Fast, accurate, and easy to use.", "positive"),
    ("This API is well-documented and reliable.", "positive"),
    ("Terrible latency, crashes frequently.", "negative"),
    ("Very confusing documentation.", "negative"),
    ("I would not recommend this library.", "negative"),
]

test = [
    ("Amazing results with minimal configuration.", "positive"),
    ("The setup process is very painful.", "negative"),
]

classifier = NaiveBayesClassifier(train)

print("Accuracy:", classifier.accuracy(test))
print("Classify new text:", classifier.classify("Incredibly fast and accurate!"))