Sentiment and thematic analysis of mobile banking app reviews for Ethiopian banks using NLP and Oracle DB – a fintech consulting project with Omega Consultancy.
Here is a comprehensive `README.md` file that a new developer can use to understand and run the project from scratch. It covers everything we did, from setting up the database to deriving the final insights.
-----
# Bank Reviews Analysis and Sentiment Analysis Project
## 📜 Project Overview
This project is a data science pipeline that collects, cleans, and analyzes customer reviews from bank mobile applications. The goal is to derive actionable insights, identify key user pain points and drivers of positive feedback, and provide data-driven recommendations for app improvement. The entire pipeline is designed to be reproducible and is built using Python, pandas, and a PostgreSQL database.
## 🛠️ Technology Stack
* **Python 3.x:** Core programming language.
* **Pandas:** For data manipulation and cleaning.
* **NLTK & Scikit-learn:** For text preprocessing, sentiment analysis, and theme identification.
* **PostgreSQL:** A relational database used for data persistence.
* **`psycopg2`:** The Python adapter for PostgreSQL.
* **Matplotlib & Seaborn:** For data visualization.
* **Jupyter Notebook:** For a step-by-step, interactive development environment.
## 📁 Project Structure
```
.
├── data/
│ ├── processed/
│ │ └── analyzed_reviews.csv
│ └── raw/
│ └──
├── notebooks/
│ ├── data_collection.ipynb
│ ├── data_cleaning.ipynb
│ ├── db_loading.ipynb
│ └──insights.ipynb
├── src/
│ └── db_loader.py
│ ├── analyzer.py
│ ├── scraper.py
│ └──preprocess.py
├── .gitignore
├── README.md
└── requirements.txt
```
## 🚀 Setup and Installation
### **1. Clone the Repository**
```bash
git clone
github.com
cd bank_reviews_analysis
```
### **2. Set up the Python Environment**
It's recommended to use a virtual environment.
```bash
# Create a virtual environment
python -m venv venv
# Activate the virtual environment
# On Windows
.\venv\Scripts\ …