Project 11

A deep neural network and two machine learning models will be built to tackle the challenge and compare different model performance. Additionally, data sampling techniques will be implemented to improve the model. As normal, split into 9 parts:

1. Business challenge
2. Data review
3. Data processing
4. DNN Model building
5. DNN Model evaluation
6. Decision Tree
7. Random forest
8. Sampling
9. Summary

1. Business challenge

Detecting fraud transactions is of great importance for any credit card company. We are tasked by a well-known company to detect potential frauds so that customers are not charged for items that they did not purchase. So the goal is to build a classifier that tells if a transaction is a fraud or not.

2. Data review

The dataset is the Kaggle Credit Card Fraud Detection dataset here. It contains two-day transactions made on 09/2013 by European cardholders. The dataset contains 492 frauds out of 284,807 transactions. Thus, it is highly unbalanced, with the positive (frauds) accounting for only 0.17%.

Looking at the data shown in the video, you may find it only contains numerical variables. Features V1, V2, … V28 are the principal components obtained with PCA transformation. The only features which have not been transformed are ‘Time’ and ‘Amount’. ‘Time’ is the seconds elapsed between each transaction and the first. ‘Amount’ is the transaction amount. ‘Class’ is the response variable with 1 as fraud and 0 otherwise.