Welcome to "Fraud Detection Project". This is the last project of the Capstone Series. One of the challenges in this project is the absence of domain knowledge. So without knowing what the column names are, we will only be interested in their values. The other one is the class frequencies of the target variable are quite imbalanced. We will implement Logistic Regression, Random Forest, Neural Network algorithms and SMOTE technique. Also visualize performances of the models using Seaborn, Matplotlib and Yellowbrick in a variety of ways. At the end of the project, we will have the opportunity to deploy our model by Streamlit API.
Our Streamlit's link is here.