🛍️ Automated Customer Reviews Analysis

This project provides a Streamlit-powered web interface for two powerful NLP tasks:

1. Review Sentiment Classification using a fine-tuned RoBERTa model.
2. Product Name Clustering using TF-IDF vectorization and KMeans clustering.

---

📌 Project Structure

---

💡 Tasks Overview

Task 1: Review Sentiment Classification

• Classifies reviews into:

  • Positive
  • Neutral
  • Negative

• Built using:

  • 🤗 Transformers (RoBERTa model fine-tuned on labeled customer reviews)
  • PyTorch for inference
  • Streamlit for the interactive UI

Task 2: Product Name Clustering

• Groups similar product names into meta-categories such as:

  • AmazonBasics Essentials
  • Kindle E-Readers
  • Alexa Devices & Accessories
  • Fire Tablets
  • Pet Supplies & Laptop Backpacks

• Built using:

  • Scikit-learn's TfidfVectorizer for feature extraction
  • KMeans clustering to group similar products

---

🖥️ Running the App

1. Clone the Repository

git clone https://github.com/yourusername/customer-review-analysis.git
cd customer-review-analysis