Sentiment Analysis of Flipkart Reviews | End-to-End Data Science Project with Python and TensorFlow

🎥 In this video, I take you through my complete data science project where I analyzed over 971 Trustpilot reviews of Flipkart!

💡 What You’ll Learn:
✅ How I scraped dynamic reviews data across 50+ pages
✅ Data cleaning techniques: handling missing values, removing URLs & punctuation, lemmatization, stopword removal
✅ Sentiment labeling with polarity scores
✅ Sentiment prediction models: XGBoost, SVM, Logistic Regression – with hyperparameter tuning
✅ Building a deep learning model using TensorFlow and Keras (CNN architecture)
✅ Dealing with imbalanced datasets using SMOTE
✅ Final model accuracy: 81% for sentiment prediction, 99% for rating prediction
✅ Deploying the deep learning model using Flask
✅ Building an interactive Power BI dashboard for sentiment distribution, top keywords in each rating, and more!

📈 Key Takeaways:
✔️ Real-world data science workflow: from web scraping to deployment
✔️ How to work with dynamic web scraping & pagination
✔️ Advanced NLP preprocessing techniques
✔️ Best practices for model evaluation and selection
✔️ Visualizing data insights with Power BI

🛠️ Technologies Used:

Python, Pandas, Scikit-Learn

TensorFlow & Keras

Flask for deployment

MongoDB for data storage

Power BI for data visualization

🔗 Check out my GitHub repo for the full code:

💬 Got questions or suggestions? Drop them in the comments – I’d love to hear from you!
👋 Like, share & subscribe if you find this video helpful for your data science journey.