Credit_Risk_Analysis

Overview

The purpose of this analysis is to create a supervised machine learning model that could accurately predict credit risk. Six different methods were utilized.

• 1. Naive Random Oversampling
• 2. SMOTE Oversampling
• 3. Cluster Centroid Undersampling
• 4. SMOTEENN Sampling
• 5. Balanced Random Forest Classifying
• 6. Easy Ensemble Classifying

Through each of these methods, the data was split into training and testing datasets. Accuracy scores, confusion matrices and classification reports were compiled in summary form.

Deliverable 1

Use Resampling Models to Predict Credit Risk

Random Oversampling

• Accuracy Score: 67.4%

SMOTE Oversampling

• Accuracy Score: 68.2%

Deliverable 2

Use the SMOTEENN algorithm to Predict Credit Risk

Undersampling Analysis

• Accuracy Score: 52.2%

SMOTEENN Analysis

• Accuracy Score: 68.1%

Deliverable 3

Use Ensemble Classifiers to Predict Credit Risk

Random Forest Analysis

• Accuracy Score: 64.8%

Easy Ensemble Analysis

• Accuracy Score: 92.3%

Summary

This analysis tries to find the best model that can detect if a loan is high-risk. Because of that, we need to find a model that lets the least amount of high-risk loans pass through undetected. That correlating statistic for this is the recall rate for high-risk. Looking through the different models, the ones that scored the highest were:

• Easy Ensemble Analysis (91%)
• SMOTEENN Analysis (76%)
• SMOTE Oversampling (70%)

While this is the most important statistic that is pulled from this analysis, another important statistic is recall rate for low-risk as it shows how many low-risk loans are flagged as high-risk. Looking through the different models, the ones that scored the highest were:

• Random Forest Analysis (100%)
• Easy Ensemble Analysis (94%)

However, one looks at the accuracy score to get a picture of how well the model performs in general. The models with the highest accuracy scores were:

• Easy Ensemble Analysis (92.3%)
• SMOTE Oversampling (68.2%)
• SMOTEENN Analysis (68.1%)

Therefore the model that is recommended by factoring multiple metrics is the Easy Ensemble Analysis.