Inspiration
Inspiration for this project is to minimize the credit card frads commitedd by the card holders. to minimize the losses of the credit card companies by using the previious data we can aslo take informed decisions to assign a creddit card to the user or not.
What it does
It takes basic information from the user and predicts the behaviour of the user and predit is he going to fullfil his payment to default it and by predicting it we can take a ddecison .
How we built it
Data- we collect data of the users who are using the credit card from all the credit data and by analysing the credit card holders behavoiur in payments we can classify the data into two types they are fraud or not fraud Data Cleaning- we use data data cleaning methods to format tthe data and drop the unwantted data Modeling:-we use basic classification models to predit the data patterns
Challenges we ran into
Data complexity formating issues
Accomplishments that we're proud of
we had build a robust model to predict credit card fraud using previous users patterns
What we learned
how to train and create model and adjusts its parameters for training
What's next for Credit Card Fraud Detection
I would like to add a multi data prediction system with a multiple data sets and multiple models and create a ensemblre model with it by taking more factors into the consideration the accuracy of the prediction increases.


Log in or sign up for Devpost to join the conversation.