How shakiliitju/credit-card-fraud-detection-using-machine-learning works
This repository documents a credit-card fraud detection project: transaction data is used to train four alternative classifiers, whose performance is evaluated and compared to identify suitable models. The notebooks are the only visible implementation entry points; the tree also contains reports and a presentation that communicate project results. No source excerpts were available, so the graph reflects only the workflow and capabilities stated in the README, not notebook-level call relationships or preprocessing details.
Main entry points: Credit Card Fraud Detection - K-Nearest Neighbor.ipynb K-nearest neighbors · Credit Card Fraud Detection - Logistic Regression.ipynb Logistic regression · Credit Card Fraud Detection - Support Vector Machines.ipynb Support vector machine · Credit Card Fraud Detection - Decision Tree.ipynb Decision tree
Components 9
Transaction data
- Kaggle dataset source
- Transaction dataset
Fraud models
- K-nearest neighbors
Credit Card Fraud Detection Using Machine Learning/Code/Credit Card Fraud Detection - K-Nearest Neighbor.ipynb - Logistic regression
Credit Card Fraud Detection Using Machine Learning/Code/Credit Card Fraud Detection - Logistic Regression.ipynb - Support vector machine
Credit Card Fraud Detection Using Machine Learning/Code/Credit Card Fraud Detection - Support Vector Machines.ipynb - Decision tree
Credit Card Fraud Detection Using Machine Learning/Code/Credit Card Fraud Detection - Decision Tree.ipynb
Evaluation and outputs
- Holdout evaluation
- Model comparison
Credit Card Fraud Detection Using Machine Learning/Report/Final Lab Report.pdf - Project presentation
Credit Card Fraud Detection Using Machine Learning/Presentation/AI-Report-Presentation.pptx
Connections 11
- Kaggle dataset source → Transaction dataset: provides
- Transaction dataset → K-nearest neighbors: trains model
- Transaction dataset → Logistic regression: trains model
- Transaction dataset → Support vector machine: trains model
- Transaction dataset → Decision tree: trains model
- K-nearest neighbors → Holdout evaluation: evaluates on
- Logistic regression → Holdout evaluation: evaluates on
- Support vector machine → Holdout evaluation: evaluates on
- Decision tree → Holdout evaluation: evaluates on
- Holdout evaluation → Model comparison: compares results
- Model comparison → Project presentation: communicates results