BaaS Risk Monitoring using Synthetic Data

Project Overview

We aim to build a machine learning-powered fraud detection and monitoring dashboard using IBM Synthetic Datasets for Core Banking and Money Laundering.

Project Structure

• app/: live dashboard code (Streamlit)
• data/: raw and processed datasets
• notebooks/: Jupyter exploration and model experiments
• scripts/: reusable code
• reports/: slides and visuals

Objective

• Detect high-risk transactions (e.g., laundering, check fraud)
• Build an interactive dashboard for transaction monitoring
• Demonstrate real-time fraud flagging capability

Tech Stack

Python, Scikit-learn, Pandas, Streamlit, GitHub, AWS

Team Members

• Delphin Kaduli
• Tycho Janssen
• Solomon Pinto.

Setup

git https://github.com/DelphinKdl/CUA-MDA-Capstone-BaaS-Risk-Monitoring.git
python -m venv .venv
source .venv/bin/activate # on mac on window is .\.venv\Scripts\activate
pip install -r requirements.txt