Credit Card Fraud Prediction

2023F-T3 BDM 3035 - Big Data Capstone Project 01
Group 1:
  Jefford Secondes
  Jovi Fez Bartolata
  Luz Zapanta
  Maricris Resma
  Keyvan Amini

PROJECT DESCRIPTION:

The major goal of the project is to develop a predictive modelling application that effectively predicts credit card fraudulent actions, such as identity theft, financial fraud, and unauthorized access. Various supervised machine learning classifier algorithms will be developed and tested upon to be able to identify the model that gives the best accuracy that satisfies the requirement of greater than 80% accuracy. Historical data will be evaluated using various machine learning models, identifying abnormalities and trends suggestive of fraudulent activities. The fraud detection application, aimed to protect the integrity of Bank of Mississauga's operations, shall be used to mitigate fraudulent transactions. This comprehensive approach addresses both known fraud trends and emergent threats, providing a proactive defense mechanism against developing fraud strategies.

OBJECTIVES:

● Retrieve and manage credit card transaction Data from Google Cloud Platform

● Create a machine learning classifier model capable of detecting credit card fraud using Logistic Regression Model, Decision Tree, or Random Forest.

● Improve the adaptability of the model to evolving fraud patterns through continuous learning.

● Reduce false positives to guarantee a seamless and trustworthy user experience.

● Deliver the pickle file that can be integrated into Bank of Mississauga’s organizational systems and databases to assure seamless operation