🧠 Sentiment Analysis on Product Reviews

This project applies Natural Language Processing (NLP) techniques to classify customer reviews from Amazon and a product's official site into Positive, Negative, or Neutral sentiments. It uses both traditional NLP methods and transformer-based deep learning models for comparison.

---

📌 Project Goals

• Clean and preprocess customer reviews using NLP techniques.
• Analyze and classify review sentiments using:
  • VADER (Lexicon-Based)
  • TextBlob
  • RoBERTa (Transformer-Based)
  • BERT
• Evaluate model performance using source ratings for validation.

---

🧰 Technologies Used

• Languages: Python
• Libraries:
  • NLP: nltk, textblob, vaderSentiment, transformers
  • ML: scikit-learn
  • Data Handling: pandas, numpy
  • Visualization: matplotlib, seaborn

---

🧠 Models Overview

Model	Type	Highlights
VADER	Lexicon-based	Best for social media-like text
TextBlob	Rule-based	Quick and lightweight sentiment analyzer
BERT	Transformer-based	Pre-trained language model from Google
RoBERTa	Transformer-based	Improved variant of BERT by Facebook

---

🧪 Preprocessing Steps

• Lowercasing and text cleaning
• Tokenization and lemmatization
• Stopwords removal
• Vectorization (TF-IDF and transformer tokenization)

---

📊 Evaluation

Sentiment predictions were cross-validated using actual star ratings to measure accuracy.
📌 RoBERTa achieved the highest performance among all models.

---

🗂️ How to Run

1. Clone the repository:

   git clone https://github.com/ShwetaPardhi0/Sentiment-Analysis-nlp.git
   cd Sentiment-Analysis-nlp