README.md

<!--
================================================================================
* Sentiment Analysis Dataset - README
*
* Project: Sentiment Analysis Dataset
* Description: Text sentiment analysis dataset with labeled reviews, comments,
* and social media posts for sentiment classification models.
* Category: Text Data
* Difficulty: Intermediate
*
* Author: Molla Samser (Founder)
* Designer & Tester: Rima Khatun
* Website: https://rskworld.in
* Email: help@rskworld.in | support@rskworld.in
* Phone: +91 93305 39277
*
* © 2026 RSK World - Free Programming Resources & Source Code
* All rights reserved.
================================================================================
-->

# 📊 Sentiment Analysis Dataset

[](https://rskworld.in)
[](https://rskworld.in)
[](https://rskworld.in)
[](https://python.org)
[](https://rskworld.in)

Text sentiment analysis dataset with labeled reviews, comments, and social media posts for sentiment classification models. Includes **Python scripts for data generation, preprocessing, analysis, visualization, and model training**.

## 🌟 Features

- ✅ **Labeled Sentiment Data** - Pre-labeled text with positive, negative, and neutral classifications
- ✅ **Multiple Text Sources** - Diverse collection from product reviews, social media, and comments
- ✅ **Training & Test Sets** - Pre-split datasets ready for ML model development
- ✅ **Preprocessed Versions** - Cleaned and tokenized data ready for NLP pipelines
- ✅ **Ready for NLP Models** - Compatible with NLTK, spaCy, and popular frameworks
- ✅ **Python Scripts** - Generate unlimited data, train models, visualize results
- ✅ **Interactive Demo** - Beautiful web interface to explore the dataset

## 📁 Project Structure

```
sentiment-analysis/
├── 📂 data/
│ ├── sentiment_data.csv # Main dataset (CSV)
│ ├── sentiment_data.json # Main dataset (JSON)
│ ├── sentiment_data.txt # Main dataset (TXT)
│ ├── train_data.csv # Training set (80%)
│ └── test_data.csv # Test set (20%)
├── 📂 preprocessed/
│ ├── cleaned_data.csv # Cleaned/normalized text
│ └── tokenized_data.json # Tokenized data for NLP
├── 📂 scripts/ # 🆕 Python Scripts
│ ├── generate_data.py # Generate unlimited synthetic data
│ ├── preprocess_data.py # Preprocess and clean text
│ ├── analyze_sentiment.py # Analyze sentiment with multiple methods
│ ├── visualize_data.py # Generate charts and visualizations
│ ├── train_model.py # Train ML models
│ └── requirements.txt # Python dependencies
├── 📂 css/
│ └── styles.css # Demo page styles
├── 📂 js/
│ └── script.js # Demo page scripts
├── index.html # Interactive demo page
├── README.md # This file
└── LICENSE # License information
```

## 🚀 Quick Start

### 1️⃣ Install Python Dependencies

```bash
cd scripts
pip install -r requirements.txt

# Download NLTK data
python -c "import nltk; nltk.download('punkt'); nltk.download('stopwords'); nltk.download('wordnet')"
```

### 2️⃣ Generate Custom Dataset

```bash
# Generate 1000 samples
python generate_data.py --samples 1000

# Generate 5000 balanced samples in all formats
python generate_data.py --samples 5000 --balanced --all-formats

# Generate with train/test split
python generate_data.py --samples 10000 --split 0.8 --output ../data/custom
```

### 3️⃣ Preprocess Data

```bash
# Basic preprocessing
python preprocess_data.py --input ../data/sentiment_data.csv

# Advanced preprocessing with lemmatization
python preprocess_data.py --input ../data/sentiment_data.csv --lemmatize --remove-stopwords

# Build vocabulary
python preprocess_data.py --input ../data/sentiment_data.csv --build-vocab
```

### 4️⃣ Analyze Sentiment

```bash
# Analyze single text
python analyze_sentiment.py --text "I love this product!"

# Interactive mode
python analyze_sentiment.py --interactive

# Evaluate on dataset
python analyze_sentiment.py --file ../data/sentiment_data.csv --evaluate
```

### 5️⃣ Visualize Data

```bash
# Generate all charts
python visualize_data.py --input ../data/sentiment_data.csv --all-charts

# Generate HTML report
python visualize_data.py --input ../data/sentiment_data.csv --html-report
```

### 6️⃣ Train ML Models

```bash
# Train Naive Bayes model
python train_model.py --input ../data/sentiment_data.csv --model naive_bayes

# Train all models and save best
python train_model.py --input ../data/sentiment_data.csv --all-models --save

# Train with custom train/test files
python train_model.py --train ../data/train_data.csv --test ../data/test_data.csv --model svm
```

## 📜 Python Scripts Reference

### `generate_data.py` - Data Generator

Generate synthetic sentiment analysis data with customizable parameters.

```bash
python generate_data.py [OPTIONS]

Options:
-n, --samples Number of samples to generate (default: 1000)
-o, --output Output file path (without extension)
-f, --format Output format: csv, json, txt, all
-b, --balanced Generate balanced dataset
-s, --split Train/test split ratio (e.g., 0.8)
-m, --include-metadata Include metadata in samples
-a, --all-formats Export in all formats
--seed Random seed for reproducibility
```

### `preprocess_data.py` - Data Preprocessor

Clean and preprocess text data for NLP models.

```bash
python preprocess_data.py [OPTIONS]

Options:
-i, --input Input file path (required)
-o, --output Output directory
-l, --lowercase Convert to lowercase
-s, --remove-stopwords Remove stopwords
-p, --remove-punctuation Remove punctuation
--lemmatize Apply lemmatization
--stem Apply stemming
-f, --extract-features Extract sentiment features
-v, --build-vocab Build vocabulary file
```

### `analyze_sentiment.py` - Sentiment Analyzer

Analyze sentiment using multiple methods (lexicon-based, VADER, TextBlob).

```bash
python analyze_sentiment.py [OPTIONS]

Options:
-t, --text Text to analyze
-f, --file File to analyze
-e, --evaluate Evaluate predictions against labels
-m, --method Analysis method: lexicon, vader, textblob, ensemble
-i, --interactive Run in interactive mode
-o, --output Output file for results
```

### `visualize_data.py` - Data Visualizer

Generate charts, word clouds, and statistical reports.

```bash
python visualize_data.py [OPTIONS]

Options:
-i, --input Input file path (required)
-o, --output Output directory for charts
-a, --all-charts Generate all available charts
-s, --stats-only Only print statistics
-r, --html-report Generate HTML report
```

### `train_model.py` - Model Trainer

Train and evaluate machine learning models for sentiment classification.

```bash
python train_model.py [OPTIONS]

Options:
-i, --input Input data file
--train Training data file
--test Test data file
-m, --model Model: naive_bayes, svm, logistic_regression, random_forest
-a, --all-models Train all available models
-s, --split Train/test split ratio
-v, --vectorizer Vectorizer: tfidf, count
-cv, --cross-validate Cross-validation folds
--save Save the best model
-o, --output Output directory for models
```

## 📊 Dataset Statistics

| Metric | Value |
|--------|-------|
| Total Samples | 50+ (expandable with generator) |
| Sentiment Classes | 3 (Positive, Neutral, Negative) |
| Text Sources | 3 (Reviews, Social Media, Comments) |
| Avg. Text Length | ~142 characters |
| File Formats | CSV, JSON, TXT |
| Languages | English |

## 🏷️ Sentiment Distribution

- **Positive**: ~37% of samples
- **Neutral**: ~30% of samples
- **Negative**: ~33% of samples

## 🛠️ Technologies & Dependencies

### Core Technologies
| Technology | Description |
|------------|-------------|
| CSV | Standard comma-separated values format |
| JSON | JavaScript Object Notation format |
| TXT | Plain text format |
| Python 3.8+ | Programming language |

### Python Libraries
| Library | Purpose |
|---------|---------|
| NLTK | Natural language processing |
| spaCy | Industrial NLP |
| TextBlob | Simple sentiment analysis |
| VADER | Sentiment analysis |
| scikit-learn | Machine learning |
| matplotlib | Data visualization |
| wordcloud | Word cloud generation |

## 📖 Usage Examples

### Loading Data in Python

```python
import pandas as pd
import json

# Load CSV
df = pd.read_csv('data/sentiment_data.csv', comment='#')

# Load JSON
with open('data/sentiment_data.json', 'r') as f:
data = json.load(f)
samples = data['data']
```

### Training a Custom Model

```python
from scripts.train_model import SentimentModelTrainer, load_data, prepare_data

# Load data
data = load_data('data/sentiment_data.csv')
texts, labels = prepare_data(data)

# Train model
trainer = SentimentModelTrainer()
trainer.train(texts[:80], labels[:80], 'svm')

# Evaluate
results = trainer.evaluate(texts[80:], labels[80:])
print(f"Accuracy: {results['svm']['accuracy']:.2%}")
```

### Interactive Sentiment Analysis

```python
from scripts.analyze_sentiment import EnsembleSentimentAnalyzer

analyzer = EnsembleSentimentAnalyzer()

# Analyze text
result = analyzer.analyze("This product is absolutely amazing!")
print(f"Sentiment: {result['ensemble']['sentiment']}")
```

## 📜 License

This dataset is provided for **educational purposes only**.

See the [LICENSE](LICENSE) file for more details.

## 👨‍💻 Author

**Molla Samser** - Founder of RSK World

- 🌐 Website: [https://rskworld.in](https://rskworld.in)
- 📧 Email: help@rskworld.in
- 📞 Phone: +91 93305 39277

### Design & Testing

**Rima Khatun** - Designer & Tester at RSK World

## 🤝 Contributing

Contributions are welcome! If you'd like to contribute:

1. Fork the repository
2. Create a new branch
3. Make your changes
4. Submit a pull request

## 📞 Contact

For questions, suggestions, or support:

- **General Inquiries**: info@rskworld.in
- **Support**: support@rskworld.in
- **Website**: [https://rskworld.in/contact.php](https://rskworld.in/contact.php)

## ⭐ Support

If you find this dataset helpful, please consider:

- ⭐ Starring this repository
- 📢 Sharing with others
- 🔗 Linking back to RSK World

---

<p align="center">
<strong>© 2026 RSK World - Free Programming Resources & Source Code</strong><br>
Founded by <strong>Molla Samser</strong> | Designed by <strong>Rima Khatun</strong>
</p>

<p align="center">
<a href="https://rskworld.in">Website</a> •
<a href="https://rskworld.in/about.php">About</a> •
<a href="https://rskworld.in/contact.php">Contact</a>
</p>