Git Product home page Git Product logo

credit_risk_analysis's Introduction

credit_banner

Credit Risk Analysis

Project Overview

The purpose of this analysis is to employ different Machine Learning techniques to train and evaluate models with unbalanced classes. Using credit card data, the goal is to determine whether or not certain machine learning models should be used to predict credit risk. The models utilized in this project are listed below:

  • Naive Random Oversampling
  • SMOTE (Synthetic Minority Oversampling Technique)
  • ClusterCentroids Undersampling
  • SMOTEENN (SMOTE and Edited Nearest Neighbors) Over-and-Under-sampling
  • Balanced Random Forest Classifier
  • Easy Ensemble AdaBoost Classifier

Resources: Credit score risk dataset, Jupyter Notebook, Python and additional libraries: Pandas, NumPy, scikit-learn, and Imbalanced-learn

Results

Naive Random Oversampling

  • Balanced accuracy score: 0.64
  • Precison: high credit risk at 0.01, low credit risk at 1.00
  • Recall (sensitivity): high credit risk at 0.62, low credit risk at 0.65

N_over

SMOTE

  • Balanced accuracy score: 0.63
  • Precison: high credit risk at 0.01, low credit risk at 1.00
  • Recall (sensitivity): high credit risk at 0.62, low credit risk at 0.64

SMOTE

ClusterCentroids Undersampling

  • Balanced accuracy score: 0.51
  • Precison: high credit risk at 0.01, low credit risk at 1.00
  • Recall (sensitivity): high credit risk at 0.59, low credit risk at 0.43

CC

SMOTEENN

  • Balanced accuracy score: 0.65
  • Precison: high credit risk at 0.01, low credit risk at 1.00
  • Recall (sensitivity): high credit risk at 0.70, low credit risk at 0.58

SMOTEENN

BalancedRandomForestClassifier

  • Balanced accuracy score: 0.79
  • Precison: high credit risk at 0.04, low credit risk at 1.00
  • Recall (sensitivity): high credit risk at 0.67, low credit risk at 0.91
  • Largest feature in this analysis was "total_rec_prncp" at 0.07.

BRF BRF_features

Easy Ensemble AdaBoost Classifier

  • Balanced accuracy score: 0.93
  • Precison: high credit risk at 0.07, low credit risk at 1.00
  • Recall (sensitivity): high credit risk at 0.91, low credit risk at 0.94

boost

Summary and Recommendation

Looking at the results from various machine learning models, the model that had the strongest predictive power in predicting credit risk (high and low) was Easy Ensemble AdaBoost Classifier. Having an accuracy score at about 93%, this model is best suited for predicting credit risk.

credit_risk_analysis's People

Contributors

nhafer88 avatar

Watchers

 avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.