Decentralized Credit Scoring with Alternative Data

This project explores a decentralized approach to credit scoring by leveraging non-traditional data sources such as phone bills, utility payments, and other financial records. Built using Python, Solidity, and the Brownie framework, the system calculates credit scores off-chain and stores summarized results on-chain using a local blockchain network powered by Ganache.

Purpose

To prototype a transparent credit scoring mechanism that:

• Uses alternative financial data to determine creditworthiness
• Stores credit score summaries on-chain for traceability
• Enables queries of credit score metadata per blockchain update

This system is aimed at individuals looking to build decentralized credit profiles using responsible financial behavior outside of traditional banking systems.

---

Learning & Development Highlights

This project was a personal deep dive into decentralized development and scripting. Key learning milestones include:

• Git and Version Control: Applied Git in a structured project setting for the first time.
• Smart Contracts with Brownie: Learned how to write, deploy, and interact with Solidity contracts through Brownie and Web3.py.
• Data Processing in Python: Built a system that recursively searches for CSV files, cleans the data, and merges them into a unified format for score processing.
• Manual Blockchain Interaction: Gained hands-on experience deploying and updating contracts via the Brownie CLI with Ganache.
• Problem Solving:
  • Faced early issues with overly strict data validation filters resulting in empty merged datasets.
  • Adapted by loosening the cleaner logic to convert malformed data into usable formats instead of discarding it outright.

---

Tech Stack

• Python – Data parsing, CSV compilation, blockchain interaction scripts
• Brownie – Smart contract development and deployment
• Solidity – Credit score contract logic
• Ganache – Local Ethereum network for development
• Web3.py – Python interface to interact with Ethereum blockchain

---

Core Functionality

1. CSV Compilation

• Scans the entire project directory for CSV files
• Cleans and merges valid entries into one master CSV
• Ensures data consistency before blockchain submission

2. Smart Contract Storage

• Stores a mapping of {userID: creditScore} on-chain
• Designed for easy updates when new credit score data is available

3. Blockchain Updates

• Manual update pipeline via the Brownie console
• update_credit_scores pushes latest scores from the combined CSV

4. Querying Data

• Smart contract exposes a function to retrieve metadata about on-chain entries and version history

Project Layout

.
├── smart_contracts/
│   ├── contracts/
│   │   └── CreditScoreContract.sol
│   ├── scripts/
│   │   ├── deploy_credit_contract.py
│   │   ├── update_credit_scores.py
│   │   ├── query_credit_scores.py
│   │   └── show_block.py
├── CreditScoreCreation.py
├── CsvCompiler.py
├── DataSetCleaning.py
├── *.csv
├── DirectoryPathways.env
└── brownie-config.yaml

Getting Started

1. Start Ganache

2. Open Brownie Console

brownie console

3. Deploy the Smart Contract

run('scripts/deploy_credit_contract')

4. Compile and Merge CSVs

python CsvCompiler.py

5. Update Credit Scores On-Chain

run('scripts/update_credit_scores')

6. Query Metadata or Scores

run('scripts/query_credit_scores')

Note: You’ll need Python, Brownie, and Ganache properly installed. See the Brownie Installation Guide for setup details.

Final Notes

• All code is thoroughly commented for ease of understanding.
• This project has no frontend interface; it operates entirely through the CLI.
• The focus is on demonstrating end-to-end credit score calculation and blockchain integration using open-source tools.

AUTHOR: Melvin Thoompunkal