🧠 AI with Python – 📈 Sentiment Analysis with TextBlob

Posted On: September 2, 2025

Description:

Understanding how people feel from text — whether it’s a review, comment, or email — is a common requirement in real projects. TextBlob offers a very quick way to get a sentiment score without heavy setup or large models.

Why TextBlob?

Simple API: Access sentiment with TextBlob(text).sentiment.
Fast baseline: Great for quick experiments or dashboards.
Readable outputs: Two numbers — polarity (−1 to 1) and subjectivity (0 to 1).

Installing the Package

pip install textblob

If you encounter lookup errors for corpora, run:

python -m textblob.download_corpora

Minimal Implementation

Compute sentiment for a single sentence:

from textblob import TextBlob

text = "I absolutely love how simple TextBlob makes sentiment analysis!"
pol, sub = TextBlob(text).sentiment.polarity, TextBlob(text).sentiment.subjectivity

Map polarity to a simple label:

def polarity_to_label(p, eps=0.05):
    return "Positive" if p > eps else "Negative" if p < -eps else "Neutral"

Sample Output

For the input:

"I absolutely love how simple TextBlob makes sentiment analysis!"

You might get:

Polarity: 0.625, Subjectivity: 0.700 → Label: Positive

Key Takeaways

Polarity tells you how negative/positive a sentence is.
Subjectivity indicates how opinionated the text is.
Quick, lightweight baseline for comments, reviews, and feedback.

Limitations

Not robust to sarcasm, irony, or complex context.
May underperform on domain-specific jargon (e.g., finance, medicine).
For multi-language support or higher accuracy, consider model-based approaches (e.g., Transformers pipelines).

Code Snippet:

# Install TextBlob (run in a notebook cell or terminal)
# !pip install textblob

# Optional: download corpora used by TextBlob (only if you see lookup errors)
# !python -m textblob.download_corpora


from textblob import TextBlob

# Sample inputs
single_sentence = "I absolutely love how simple TextBlob makes sentiment analysis!"
texts = [
    "The movie was fantastic and truly inspiring.",
    "It was okay, nothing special.",
    "This is the worst service I have ever experienced."
]


# Create a TextBlob and read sentiment
blob = TextBlob(single_sentence)
polarity = blob.sentiment.polarity      # how positive/negative
subjectivity = blob.sentiment.subjectivity  # how subjective/objective

print("Sentence:", single_sentence)
print("Polarity:", round(polarity, 3))
print("Subjectivity:", round(subjectivity, 3))


def polarity_to_label(p, eps=0.05):
    """
    Map polarity score to a label:
    - p > +eps → 'Positive'
    - p < -eps → 'Negative'
    - otherwise → 'Neutral'
    """
    if p > eps:
        return "Positive"
    if p < -eps:
        return "Negative"
    return "Neutral"


for t in texts:
    p = TextBlob(t).sentiment.polarity
    s = TextBlob(t).sentiment.subjectivity
    print(f"\nText: {t}")
    print(f"Polarity: {p:.3f}  |  Subjectivity: {s:.3f}  |  Label: {polarity_to_label(p)}")

← →	move
↑	rotate
↓	soft drop
Space	hard drop
P	pause / resume

🧠 AI with Python – 📈 Sentiment Analysis with TextBlob

Description:

Why TextBlob?

Installing the Package

Minimal Implementation

Sample Output

Key Takeaways

Limitations

Code Snippet:

Comments

Add Your Comment

🧠 AI with Python – 📈 Sentiment Analysis with TextBlob

Description:

Why TextBlob?

Installing the Package

Minimal Implementation

Sample Output

Key Takeaways

Limitations

Code Snippet:

Comments Show Comments

Add Your Comment

Related Posts

🧠 AI with Python – 🐍🛍️📊 Clustering Customers using K-Means

⚡️ Saturday AI Sparks 🤖 - 🖼️📝 Image Captioning with Pre-trained Model

🧠 AI with Python – 🖼️🔢 Visualise Digits using matplotlib

Comments