Credit card fraud is a global problem, with 32 billion dollars in 2021. Existing solutions used by banks rely on rule-based engines, where rules used to detect fraud have to be manually added, and fraud is only detected after it happens. Machine learning offers a promising solution to this, having many advantages compared to more traditional rule-based engines, and in recent years, there has been a lot of research on applying machine learning to this problem.
In this project, we develop three fraud detectors based on three machine learning algorithms: Deep Neural Network, Random Forest, and Naïve Bayes, with the goal of detecting fraud in real-time with a minimal number of false positives.
Explore the docs »
View Demo
·
Report Bug
·
Request Feature
Table of Contents
Credit card fraud is a global problem, with 32 billion dollars in 2021. Existing solutions used by banks rely on rule-based engines, where rules used to detect fraud have to be manually added, and fraud is only detected after it happens. Machine learning offers a promising solution to this, having many advantages compared to more traditional rule-based engines, and in recent years, there has been a lot of research on applying machine learning to this problem.
In this project, we develop three fraud detectors based on three machine learning algorithms: Deep Neural Network, Random Forest, and Naïve Bayes, with the goal of detecting fraud in real-time with a minimal number of false positives.
The models are trained and evaluated with a data set from Kaggle Credit Card Fraud Detection. The data set containes ~284 000 anonymized credit card transactions labeled as fraudulent or genuine, where only 0.127% are fraudulent.
To get a local copy up and running follow these simple example steps.
- Download the data set from Kaggle Credit Card Fraud Detection.
- Clone the repo
git clone https://github.com/sigurdurb/credit-card-fraud-detector.git
- Install required packages
pip install -r requirements.txt
- Playaround as you wish
python evaludate.py
This Project was developed by the Fraud Prevention Force
- Eva Ósk EfaOSk
- Guðmundur Óli gudmunduro
- Sigurður sigurdurb
Project Link: https://github.com/sigurdurb/credit-card-fraud-detector