📝 Sentiment Analysis of Reviews Using TextBlob

📖 Overview

This project implements sentiment analysis on customer reviews using the TextBlob library in Python.
The goal is to analyze textual data from a dataset of reviews and classify their sentiments into categories like:

✅ Very Negative
✅ Negative
✅ Neutral
✅ Positive

It also includes visualizations to represent the sentiment distribution using heatmaps.

---

✨ Features

• 🔍 Sentiment Analysis: Computes sentiment polarity scores for each review using TextBlob.
• 🧹 Data Preprocessing: Cleans and prepares text data by removing stopwords and unnecessary tokens.
• 📊 Visualization: Generates heatmaps to visually display the distribution of sentiment scores across categories.
• 📁 CSV Input/Output: Reads from a CSV file containing cleaned reviews and writes the sentiment results to a new CSV file.

---

⚙️ Requirements

You’ll need the following Python libraries:

• pandas
• textblob
• seaborn
• matplotlib

Install them using:

pip install pandas textblob seaborn matplotlib

---

📂 Dataset

Prepare a CSV file named (for example):

cleaned_reviews.csv

with at least one column:

Review
The product was excellent and arrived early!
Terrible customer service, very disappointed.

---

🚀 Usage

1️⃣ Run the main script:

python sentiment_analysis.py

2️⃣ The script will:

• Read the reviews from the CSV file.
• Calculate sentiment polarity scores.
• Classify each review into a sentiment category.
• Generate visualizations (like heatmaps).
• Write the results to a new CSV file (e.g., sentiment_results.csv).

---

📊 Sentiment Categories

The polarity scores from TextBlob are mapped as:

Polarity Score Range	Category
-1.0 to -0.6	Very Negative
-0.6 to -0.2	Negative
-0.2 to 0.2	Neutral
0.2 to 1.0	Positive

---

🛠 Example Code Snippet

from textblob import TextBlob

def get_sentiment_category(polarity):
    if polarity <= -0.6:
        return "Very Negative"
    elif polarity <= -0.2:
        return "Negative"
    elif polarity <= 0.2:
        return "Neutral"
    else:
        return "Positive"

blob = TextBlob("The product was excellent and arrived early!")
polarity = blob.sentiment.polarity
category = get_sentiment_category(polarity)

print(f"Polarity: {polarity}, Category: {category}")

---

📈 Visualizations

The script uses seaborn and matplotlib to create heatmaps showing the sentiment distribution across your dataset.

---

📜 License

This project is open-source under the MIT License.

---

🙏 Acknowledgments

Special thanks to:

• TextBlob: For its straightforward and effective sentiment analysis tools.
• Seaborn and Matplotlib: For making data visualization easy and beautiful.
• Pandas: For simplifying data handling.

---

💡 Feel free to fork this project and customize it for your own review datasets!