Data Breakdown

Feature Importance

GGRAPH Visualization

Visualize Feature Importance

Correlation Plot

XGBoost

[1] "This XGBoost took 4.98 Minutes to Run"
Confusion Matrix and Statistics

          Reference
Prediction  0  1  2  3  4  5
         0 91  0  0  0  0  0
         1  1 56  7  0  0  0
         2  0 12 75  1  0  0
         3  0  2  0 57  1  1
         4  0  0  0 21 52 13
         5  0  0  0  0  1 50

Overall Statistics
                                               
               Accuracy : 0.8639               
                 95% CI : (0.8284, 0.8945)     
    No Information Rate : 0.2086               
    P-Value [Acc > NIR] : < 0.00000000000000022
                                               
                  Kappa : 0.8363               
 Mcnemar's Test P-Value : NA                   

Statistics by Class:

                     Class: 0 Class: 1 Class: 2 Class: 3 Class: 4 Class: 5
Sensitivity            0.9891   0.8000   0.9146   0.7215   0.9630   0.7812
Specificity            1.0000   0.9784   0.9638   0.9890   0.9121   0.9973
Pos Pred Value         1.0000   0.8750   0.8523   0.9344   0.6047   0.9804
Neg Pred Value         0.9971   0.9629   0.9802   0.9421   0.9944   0.9641
Precision              1.0000   0.8750   0.8523   0.9344   0.6047   0.9804
Recall                 0.9891   0.8000   0.9146   0.7215   0.9630   0.7812
F1                     0.9945   0.8358   0.8824   0.8143   0.7429   0.8696
Prevalence             0.2086   0.1587   0.1859   0.1791   0.1224   0.1451
Detection Rate         0.2063   0.1270   0.1701   0.1293   0.1179   0.1134
Detection Prevalence   0.2063   0.1451   0.1995   0.1383   0.1950   0.1156
Balanced Accuracy      0.9946   0.8892   0.9392   0.8552   0.9376   0.8893

XGBoost has an accuracy of 86.4%.

XGBoost Feature Importance: Cover, Frequency, Gain




PCA Clustering