Credit-Risk-Analysis

A project for the Neural Networks discipline mentored by Prof. Dr. Germano Vasconcelos (http://www.cin.ufpe.br/~gcv/web_lci/germano.html) at Centro de Informática, UFPE, Brazil (cin.ufpe.br). Using mlp to classifier samples as either good or bad in credit risk analysis context.

Stack used in this project

1. numpy
2. pandas
3. keras (using MLP model)
4. matplotlib

Dataset

Link to folder

Some Results

To achieve good results different mlps configurations were tested