/heloc-machine-learning

Heloc Prediction Project is a data science and machine learning project to make creditworthiness predictions that a bank might use in order to create a favorable outcome in the loaning process

Primary LanguageJupyter NotebookMIT LicenseMIT

Heloc Predictor Project

This project focus on creating a detail analysis and machine learning for decision making predictions based on MSTZ Heloc Dataset huggingface.co.

Purpose of the project

Heloc lending process improvement is becoming a standard practice for financial institutions. With a precise understanding of the data and given predictions on loan approvals, the bank can improve its decision-making process to minimize risk and streamline approvals. We decided to engage in this project because we believe it's an excellent opportunity and a real-world example to enhance our data science skills with tangible outcomes.

Team

Name Description Github Account LinkedIn
Santiago Paiz Software Engineer & Data Scientist @aaoeclipse santiago-paiz-7b2a7268
Alex "Andru" Andrushevich Software Engineer & Data Scientist @QuantGeekDev alex-andrushevich-5544845a

Installation

  1. Download MSTZ Heloc Dataset into a data/ folder in the root of the project
  2. Install python dependencies
python3 -m pip install -r requirements.txt

Setup

  1. Follow the steps in the installation
  2. setup mlflow with the correct database
mlflow ui --backend-store-uri sqlite:///mlflow.db
  1. Run the analysis.ipynb