Detecting Credit Card Fraud

Description
  • Date: December 31, 2023
  • Categories: Deep LearningEDAMachine LearningPython

View on Github

Overview

This project applies machine learning to detect credit card fraud using a Kaggle dataset. It includes data preprocessing, exploratory analysis, and model training and evaluation.

Technologies

  • Python, Pandas, NumPy
  • Matplotlib, Seaborn
  • Scikit-learn, XGBoost, PyTorch

Objectives

  • Perform EDA on the credit card transaction dataset.
  • Develop and assess Logistic Regression, XGBoost, and a custom MLP model using PyTorch.
  • Compare model performances using metrics like accuracy and F1 score.

Learning Outcomes

  • Enhanced data preprocessing and analysis skills.
  • Experience in implementing various machine learning models.
  • Insights into model evaluation and comparison in fraud detection.