Table 6 Comparing the proposed model with recent literature
Research work | Algorithms considered | Dataset used | Highest accuracy (%) | Precision (%) | Recall (%) | F1-score (%) | Specificity (%) | AUC/-ROC (%) | Negative predicted values (%) | MCC (%) | False-positive rate | False-negative rate | False discovery rate | Misclassification rate | Run time (mins.) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Amin et al. [28] | LR, KNN, RF, SVM, MLP, voting | ILPD | 88.10 with RF | 85.33 | 92.3 | 88.68 | - | 88.20 | 93.00 | - | - | 0.0700 | 0.1467 | 0.1190 | 3.337 |
Afrin et al. [29] | LR, DT, RF, AdaBoost, KNN, LDA, GB, SVM | ILPD | 94.29 with DT | 92 | 99 | 96 | - | - | 99.00 | - | - | 0.0100 | 0.0800 | 0.0571 | - |
Bulucu et al. [36] | RF, J48, GB, AdaBoost, LGBM | UCI dataset | 98.8 with LGBM | 98.1 | 99.4 | - | - | - | Â | - | - | 0.0060 | 0.1960 | 0.0120 | - |
Dritsas and Trigka [30] | NB, SVM, LR, ANN, kNN, J48, RF, RT, RepTree, RotF, AdaBoostM1, stacking, bagging, voting | ILPD | 80.1 | 80.4 | 80.1 | 80.1 | - | 88.4 | 80.10 | - | - | 0.1990 | 0.1960 | 0.1990 | - |
Gupta et al. [40] | LR, DT, RF, KNN, XGB, LGB | ILPD | 63 with LGB and RF both | 64 with RF | 63 with RF | 63 with LGB and RF both | - | - | 63.00 | - | - | 0.3700 | 0.3600 | 0.3700 | - |
Hameed et al. [41] | RF, SVM, LR, DT, AdaBoost, GB, KNN | ILPD | 80.30 with RF | 80.30 | 80.26 | 80.20 | - | - | 80.30 | - | - | 0.1974 | 0.1970 | 0.1970 | - |
Kuzhippallil et al. [32] | LR, kNN, DT, RF, GB, AdaBoost, XGB, LGBM, stacking | ILPD | 86 XGB, LGBM | 86 XGB | 86 XGB | 86 XGB | - | - | 86.00 | - | - | 0.1400 | 0.1400 | 0.1400 | 0.191 |
Nahar et al. [31] | AdaBoost, LogitBoost, RF, bagging (RepTree and J48) | ILPD | 71.53 with LogitBoost | 83.60 | 56.45 | - | - | 72.20 | 56.45 | - | - | 0.4355 | 0.1640 | 0.2847 | - |
Naseem et al. [33] | A1DE, MLP, NB, kNN, SVM, J48, CHIRP, CDT, forest-PA, RF | BUPA | 72.17% with RF | 62.07 | 68.7 | 65.22 | 74.30 | - | 79.50 | 42.28 | 0.2570 | 0.3130 | 0.3793 | 0.2783 | - |
SanikaVT | 71.36% with SVM | 100 | 71.36 | 83.28 | - | - | 0 | - | - | 0.2864 | 0 | 28.64 | - | ||
Quadir et al. [34] | GB, XGB, bagging, RF, ET, stacking | ILPD | 93.15 with stacking | 80.76 | 94.59 | 87.13 | 74.22 | 84.41 | 94.59 | - | 0.2578 | 0.0541 | 0.1924 | 0.0685 | - |
Dalal [35] | Hybrid XGB | ILPD | 93.65 | - | - | - | - | 98.70 | - | - | - | - | - | 0.0635 | - |
Zhao et al. [42] | SVM, GP, RF, XGB, bagging | BUPA | 80.35 with RF | 68.75 | 38.82 | 49.62 | - | - | 38.82 | - | - | 0.6118 | 0.3125 | 0.1965 | - |
Our paper | XGB, LGBM, GB, BDT, RF, ET, LR, DT, SVM | LDPD | 98.80 with GB | 98.50 | 98.50 | 98.50 | 97.74 | 98.60 | 98.10 | 97.08 | 0.0226 | 0.0077 | 0.0091 | 0.0120 | 4.792 |