End-to-End Bank Fraud Detection
-
- 0 Rating
- 0 Reviews
- 2 Students Enrolled
End-to-End Bank Fraud Detection
This project focuses on building a sophisticated, end-to-end machine learning pipeline to detect fraudulent financial transactions.
-
- 0 Rating
- 0 Reviews
- 2 Students Enrolled
Requirements
- Logistic Regression, Random Forest, XGBoost, Scikit-Learn, Streamlit etc.
General Overview
Description
This project, titled "End-to-End Fraud Detection," was developed at Humboldt-Universität zu Berlin in collaboration with theIDA.net and Quantinar.com, Led by Gurpreet Singh under the supervision of Prof. Dr. Wolfgang Karl Härdle, the initiative seeks to replace rigid, legacy fraud-detection systems with a dynamic Machine Learning (ML) diagnostic framework. By moving away from binary "yes/no" rules, the project provides a probability-based approach to identifying bad actors in financial systems.
The project culminated in the delivery of a Streamlit application designed for real-time use by bank tellers and fraud investigators. This tool empowers teams with actionable insights, drastically reducing investigation costs and protecting revenue while ensuring a frictionless experience for legitimate customers.
Recommended for you
Meet the instructors !
