The Bank Statement Parser is an automated tool designed to extract, clean, and categorize financial data from bank statements in various formats (PDF, CSV, XLSX). The parser uses OCR (Optical Character Recognition) and text processing techniques to accurately identify transaction details such as date, description, amount, and balance. It currently supports one bank statement template (Opay - A fintech in Nigeria) and applies rule-based classification to tag transactions (e.g., groceries, utilities, salary, etc.).
- 📄 Multi-format Support – Import statements in PDF, CSV, or XLSX formats
- 🧠 PDF Parsing – Extract data from scanned PDFs using Tesseract OCR or PDF parsing tools like pymupdf etc.
- 🧹 Data Cleaning & Normalization – Standardize transaction details (date, amount, description, balance)
- 🏷️ Transaction Categorization – Classify expenses into categories (e.g., groceries, rent, utilities)
- 🔍 Duplicate Detection – Prevent duplicate transactions during bulk imports
- 📤 Export Options – Output parsed data to JSON, CSV, or a database
BANK_STATEMENT_PARSER/
├── .venv/ # Virtual environment (not included in repo)
├── bank_parser/ # Core parsing and utilities
│ └── utils.py # Utility functions
│ ├── logger.py # Logging setup
│ ├── db_loader.py # DB loader module
│ ├── model.py # Data models
│ ├── wrangler.py # Data wrangling logic
│ └── opay_bs/ # Opay-specific logic
│ ├── data/ # Sample or raw data for Opay parsing
│ ├── __init__.py
│ ├── opay_bankstatement.pdf # Example bank statement
│ └── opay_bs_parser.ipynb # Jupyter notebook for Opay parser
├── tests/ # Unit and integration tests
│ └── __init__.py
├── main.py # Entry point for CLI
├── .gitignore # Git ignore rules
├── poetry.lock # Dependency lock file
├── pyproject.toml # Project metadata and dependencies
└── README.md # Project documentation
| Date | Description | Amount | Balance |
|---|---|---|---|
| 2025-06-01 | Tesco Supermarket | -45.80 | 1,954.20 |
| 2025-06-03 | Salary Credit | +2,000.00 | 3,954.20 |
- Python – Core logic and parsing
- Pandas – Data manipulation
- PyPDF2 / pdfplumber – PDF parsing
- Regex – Pattern-based data extraction
- PostgreSQL – Data storage
git clone https://github.com/Iyanuvicky22/bank_statement_parser.git
cd bank-statement-parserpython parser.py --file path/to/statement.pdf- Parsed transactions saved as JSON, CSV, and loaded to a connected database.
- Optional logs and error reports for failed lines or unsupported formats.
- Personal finance tracking
- Small business expense analysis
- Preprocessing for accounting/loan systems
- Budget planning and audit readiness
Contributions are welcome! Please fork the repo, create a feature branch, and submit a pull request.