Table 6.
Comparison of the diabetes detection model outcomes with the previous studies.
| Approach | Train Test Split | Result (%) | Ref. | |||
|---|---|---|---|---|---|---|
| Decision tree Random forest Naive Bayes |
70:30 train test ratio | DT | RF | NB | [13] | |
| Accuracy Precision Sensitivity Specificity F1 score AUC |
74.78 70.86 88.43 59.63 78.68 78.55 |
79.57 89.40 81.33 75.00 85.17 86.24 |
78.67 81.88 86.75 63.29 84.24 84.63 |
|||
| RF AdaBoost Soft voting classifier |
70:30 train test ratio |
RF | Ada | Voting classifier | [10] | |
| Accuracy Precision F1 score Recall AUC |
77.48 71.21 64.38 58.75 78.10 |
75.32 68.25 60.13 53.75 74.98 |
79.08 73.13 71.56 70.00 80.98 |
|||
| RF | Not mentioned | RF | ANN | K mean clustering |
[2] | |
| Accuracy AUC |
74.70 80.60 |
75.70 81.60 |
73.60 - |
|||
| ANN XGB |
Not mentioned | ANN | XGB | [12] | ||
| Accuracy Sensitivity Specificity AUC |
71.35 45.22 85.20 65.00 |
78.91 59.33 89.40 88.00 |
||||
| Naive Bayes SVM DT |
10-fold Cross-validation |
NB | SVM | DT | [11] | |
| Precision Recall F1 score Accuracy |
75.9 76.3 76 76.3 |
42.4 65.1 51.3 65.1 |
73.50 73.80 73.60 73.82 |
|||
| Proposed soft voting classifier (XgBoost + RF) | 5 fold Cross-validation |
Accuracy Precision Recall F1 score AUC |
90 88 89 95 95 |
- | ||