wakgari-Elias/fintech-review-analytics

Fintech Review Analytics — Google Play Store Customer Experience Analysis A complete end-to-end data analytics project analyzing user experience for three major Ethiopian banking apps (CBE, BOA, Dashen). This project scrapes Google Play Store reviews, preprocesses the data, performs sentiment and thematic NLP analysis, stores resultsin PostgreSQL # fintech-review-analytics # FinTech Review Analytics A project to collect, preprocess, and analyze user reviews from banking apps on the Google Play Store. This project enables insights into customer satisfaction, pain points, and potential app improvements. --- ## Project Overview **Goal:** Collect and analyze user feedback from three Ethiopian banks’ mobile apps—CBE, BOA, and Dashen Bank—to uncover trends in customer sentiment and common themes. **Banks & Apps:** | Bank | App ID | Google Play Link | |--------|--------|-----------------| | CBE | com.combanketh.mobilebanking | Link | | BOA | com.boa.boaMobileBanking | Link | | Dashen | com.dashen.dashensuperapp | Link | --- ## Task-1: Data Collection and Preprocessing ### 1. Web Scraping - Uses `google-play-scraper` to collect reviews, ratings, review dates, and app names. - Target: ≥400 reviews per bank (1,200+ total). - Saved to `data/raw/raw_reviews.csv`. ### 2. Preprocessing - Remove duplicates and missing values. - Filter only English reviews. - Normalize dates to `YYYY-MM-DD`. - Save cleaned dataset to `data/cleaned/clean_reviews.csv`. --- ## Project Structure ``` fintech-review-analytics/ │ ├─ src/ │ ├─ scrape_reviews.py # Scrapes Google Play reviews │ └─ preprocess.py # Cleans and preprocesses scraped data │ ├─ data/ │ ├─ raw/ # Raw scraped reviews │ └─ cleaned/ # Cleaned, preprocessed reviews │ ├─ .gitignore ├─ requirements.txt └─ README.md ```` --- ## Usage 1. **Set up virtual environment (Python 3.12.10)** ```bash python -m venv .venv source .venv/Scripts/activate # Windows # or source .venv/bin/activate # Mac/Linux ```` 2. **Install dependencies** ```bash pip install -r requirements.txt ``` 3. **Scrape reviews** ```bash python -m src.scrape_reviews ``` 4. **Preprocess reviews** ```bash python -m src.preprocess ``` --- ## Continuous Integration (CI) * Uses **Flake8** for Python linting. * Code is formatte …