Inspiration
Financial institutions need robust methods to prevent fraud, comply with KYC/AML regulations, and maintain an immutable record of transactions and risk scores.
What it does
Combine blockchain for immutability and machine learning for real-time risk assessment:
Smart Contracts store transaction data, KYC verification statuses, and ML-generated risk scores. ML Engine (Python) computes fraud or credit risk probabilities using enhanced features and alternative algorithms. API Layer (Flask) orchestrates requests, calls the ML model, and records data on-chain.
How we built it
Mainly used GPT for almost all brainstorming and code generation as well as technical writing structure + documentation. (Nearly 100%)
Challenges we ran into
You need 0.001 ETH to get "play" money
Accomplishments that we're proud of
Integrating and training a full neural network
What we learned
What's next for NUSF Blockchain + ML API for Finance
Deployment via Docker or Kubernetes
Log in or sign up for Devpost to join the conversation.