rakazk/google-play-review-intelligence

Google Play Review Intelligence Dashboard

A Python and Streamlit-based app for collecting Google Play Store reviews, building a review dataset, analyzing sentiment, visualizing product intelligence charts, comparing multiple apps, and exporting CSV files for research or business analysis.

Main Features

• —Collect reviews from one or multiple Google Play apps.
• —Accept Google Play package IDs or Play Store URLs.
• —Select country, language, review sort order, number of reviews, and star rating filter.
• —Build a reusable dataset during the active session.
• —Upload existing CSV datasets for analysis.
• —Analyze sentiment using VADER.
• —Visualize sentiment distribution, rating distribution, sentiment-versus-rating, daily review volume, monthly sentiment trend, keyword frequency, TF-IDF terms, word cloud, and app-to-app comparison.
• —Export newly collected, filtered, and analysis-ready datasets.

Files Required for Hugging Face Spaces

.
├── app.py
├── Dockerfile
├── requirements.txt
├── README.md
├── .dockerignore
├── .gitignore
├── .streamlit/
│   └── config.toml
├── data/
│   └── .gitkeep
└── exports/
    └── .gitkeep

Deploy on Hugging Face Spaces

1. 1.Create or open your Hugging Face account.
2. 2.Go to Spaces and click Create new Space.
3. 3.Select:
4. 4.SDK: Docker
5. 5.Visibility: Public, Private, or Protected according to your requirement
6. 6.Space name: for example, google-play-review-intelligence
7. 7.Upload all project files to the Space repository.
8. 8.Hugging Face will automatically build the Docker image and run the app on port 8501.

Deploy by Git

git clone https://huggingface.co/spaces/YOUR_USERNAME/YOUR_SPACE_NAME
cd YOUR_SPACE_NAME

# Copy this project's files into the cloned folder, then run:
git add .
git commit -m "Deploy Google Play review intelligence Streamlit app"
git push

Local Installation

python -m venv .venv
source .venv/bin/activate   # macOS/Linux
# .\.venv\Scripts\activate  # Windows PowerShell

pip install -r requirements.txt
streamlit run app.py

Example App IDs

com.whatsapp
com.instagram.android
com.facebook.katana
com.spotify.music

Recommended Workflow

1. 1.Open Collect Reviews.
2. 2.Enter one or multiple app IDs.
3. 3.Select language, country, review sort, star filter, and maximum reviews per app.
4. 4.Click Fetch Reviews.
5. 5.Open Load Dataset and select session data, uploaded CSV, or master dataset.
6. 6.Open Analyze Dashboard.
7. 7.Export the final CSV dataset for research, reporting, or machine learning.

Hugging Face Runtime Note

On Hugging Face Spaces, local files are not guaranteed to persist after a restart unless persistent storage, a Storage Bucket, or an external database is attached. For safety, always download the CSV dataset after collection.

Suggested Research Extensions

• —Add transformer-based sentiment models such as BERT or RoBERTa.
• —Add topic modeling using BERTopic or LDA.
• —Add complaint category classification.
• —Add app version-based sentiment tracking.
• —Add competitor benchmarking reports in PDF format.