📊 Python and AI - Youtube Comments Sentiment Analysis

🚀 Python AI Sentiment Analysis is a powerful AI-driven project that extracts comments from YouTube videos and classifies their sentiment using three different models: BERT, RoBERTa, and OpenAI's GPT models. This allows a deeper comparison between models and provides valuable insights into audience reactions.

---

📌 Key Features

✅ Scrapes YouTube comments via API

✅ Applies sentiment analysis with BERT, RoBERTa, and OpenAI GPT

✅ Compares results using Confusion Matrices

✅ Generates insightful visualizations

---

🛠 How It Works?

The project follows this pipeline:

1️⃣ Extract comments from YouTube using the YouTube API.
2️⃣ Process comments using Natural Language Processing (NLP).
3️⃣ Apply sentiment classification using BERT, RoBERTa, and OpenAI's GPT.
4️⃣ Compare results using confusion matrices and statistical analyses.
5️⃣ Generate visualizations to better understand sentiment distributions.

---

📂 Project Structure

📦 python_ai_sentiment_analysis
 ┣ 📜 youtube_sentiment_analysis.ipynb   # Main script
 ┣ 📜 requirements.txt                   # Project dependencies
 ┣ 📜 README.md                          # Documentation

---

🚀 Installation & Setup

1️⃣ Clone the Repository

git clone https://github.com/RenanBjj/Python-AI-Sentiment-Analysis.git
cd Python-AI-Sentiment-Analysis

2️⃣ Install Dependencies

pip install -r requirements.txt

3️⃣ Configure API Keys

Create a .env file and add your credentials:

YOUTUBE_API_KEY=your_youtube_api_key
OPENAI_API_KEY=your_openai_api_key

4️⃣ Run the Analysis

python youtube_sentiment_analysis.py

---

📊 Results & Insights

🔹 Sentiment Distribution

The system generates bar charts displaying the distribution of sentiments across different models.

🔹 Confusion Matrices

We compare how well the models align using heatmaps that illustrate model agreement and divergence.

Example of Confusion Matrix for BERT vs OpenAI:

plt.figure(figsize=(8, 6))
sns.heatmap(conf_matrix, annot=True, fmt='d', cmap="Blues", linewidths=0.5, linecolor='gray')
plt.title("Confusion Matrix: BERT vs OpenAI Sentiments")
plt.xlabel("Sentiment OpenAI")
plt.ylabel("Sentiment BERT")
plt.show()

---

🤖 Technologies Used

🔹 Python 3.8+

🔹 Pandas & NumPy - Data handling and manipulation

🔹 Matplotlib & Seaborn - Data visualization

🔹 Transformers (Hugging Face) - BERT & RoBERTa models

🔹 OpenAI API - Sentiment classification via ChatGPT

🔹 Google YouTube API - Fetching YouTube comments

---

🤝 Contribute

Want to enhance this project? Here’s how you can help:

✅ ⭐ Star this repository to show support!

✅ 💡 Open an issue with suggestions or improvements.

✅ 📌 Submit a pull request with new features or optimizations.

---

📜 License

This project is open-source under the MIT License. Feel free to use, modify, and share! 🚀

---

👨‍💻 Developed by Renan Marques - Bringing AI to Data! 🚀