🧠 RoBERTa-Based Sentiment Analysis

A simple command-line tool for performing sentiment analysis using the pretrained cardiffnlp/twitter-roberta-base-sentiment model from Hugging Face. Optimized for analyzing social media text, short reviews, and comments.

---

📌 Features

• Utilizes state-of-the-art pretrained RoBERTa transformer
• Handles noisy text with user mentions and URLs
• Predicts sentiment as Positive, Neutral, or Negative
• Outputs probability scores for interpretability

---

🔍 How It Works

1. Input: User provides a comment or review via terminal
2. Preprocessing: Cleans mentions and links
3. Encoding: Uses RoBERTa tokenizer
4. Prediction: Model outputs class scores processed via softmax
5. Output: Displays score per sentiment and predicted class

---

🧪 Example

Enter the Comment Or Review  : I love this product! Works perfectly.
Negative 0.02
Neutral 0.15
Positive 0.83
The Sentiment of the Comment is : Positive