Credit Card Default Prediction

Machine Learning Neutral Network Model

Get Started

Based on the 24 variables, we created a regression model to predict a customer's default likehood.

The dataset contains information from more than 30,000 customers in Taiwan. The currency is Taiwan dollar (TN$).


A Look at the Dataset

Customer's Age

An integer variable to indicate the age

Customer's Gender

Female or male

Credit Card Limit Balance

The maximum amount of the cutomer's credit card line

Education Level

The highest level of education a customer received

Billing and Payment Amounts

We collected billing and payment amounts for the last 6 months

Other Variables

The last 6 months of payment past due statuses

How the dataset looks

See Visualization Charts

Machine Learning

Create the prediction model

Predict the Results

Variables We Input

About the Program

USC Data Science BootCamp 2018

About the Team