A sentiment analysis project on user reviews of the Gojek mobile application (version 5.18.2) from Google Play Store. This project combines data scraping, preprocessing, machine learning modeling, and web-based visualization to deliver actionable insights about user feedback. The application features an interactive dashboard for sentiment summary and detailed analysis.
| Category | Stack |
|---|---|
| Data Analysis | Python (Pandas, NumPy) |
| Machine Learning | Scikit-learn (TF-IDF, Classifiers), NLTK (optional) |
| Visualization | Matplotlib, Seaborn |
| Web Interface | Flask, Jinja2 Templates, Bootstrap |
| Deployment | Streamlit (Optional for Demo) |
| Dataset Source | Google Play Reviews (Gojek v5.18.2) |
- Google Play review scraping & preprocessing (cleaning, tokenizing).
- Sentiment classification (positive, neutral, negative) using supervised learning models.
- Visualization of sentiment distribution and keyword frequency.
- Web-based dashboard for interactive exploration of analysis results.
- API endpoint integration for future scalability.
You can access the deployed web application at: https://gojek-sentiment.dodevca.com
- Clone this repository.
- Create and activate a virtual environment.
python3 -m venv venv source venv/bin/activate # For Linux/MacOS # OR venv\Scripts\activate # For Windows
- Install required dependencies.
pip install -r requirements.txt
- Run the Jupyter Notebook for data preprocessing & model training.
jupyter notebook NLP.ipynb
- To run the web application (Flask):
python app.py
- Open your browser and go to
http://localhost:5000.
- Integrate more advanced models (BERT, LSTM) for enhanced prediction accuracy.
- Add real-time review fetching using Google Play API/Scraper.
- Expand dataset to include version-wise sentiment comparison.
- Deploy as a scalable microservice API for sentiment analysis.
This project is collaboratively developed by:
Interested in collaborating or enhancing this project? Reach me at LinkedIn or visit dodevca.com.
Initiated by Dodevca & Team, open for collaboration and continuous refinement.