Proposed by | Used dataset | Feature selection | Algorithm | Accuracy |
---|---|---|---|---|
Khan et al. (2018) | UNSW-NB 15 dataset | Feature importance (RF) | XGBoost, RF, Bagging, KNN, DT | 71.43%, 74.87%, 74.64%, 74.22%, 71.10% |
Tama and Rhee (2019) | NSL KDD, UNSW-NB 15 and GPRS | Complete feature | GBM | 91.31% (UNSW-NB), 91.82% (KDDTest +), 86.51% (KDDTest-21) |
Jing and Chen (2019) | UNSW-NB 15 | All features | SVM | 85.99% (binary classification), 75.77% (multiclass classification) |
Kasongo and Sun (2020) | UNSW-NB 15 | XGBoost algorithm | SVM, Logistic Regression, KNN, DT, ANN | 60.89, 77.64, 84.46, 90.85, 84.39 |
Ingre and Yadav (2015) | NSL-KDD | filter-based | DT | 83.66% (multiclass classification), 90.30% (binary classification) |
Osanaiye et al. (2016) | NSL-KDD | Chi-Square, information gain, gain ratio, and relieff algorithm | DT | 99.67% |
Alazzam et al. (2020) | UNSW-NB 15, NSL-KDD, KDDCup99 | Sigmoid PIO Cosine PIO | DT DT | 91.7% (UNSW-NB 15); 88.3% (NSL-KDD); 96% (KDDCup99); 91.3% (UNSW-NB) 86.9% (NSL-KDD) 94.7% (KDDCup99) |
Aboueata et al. (2019) | UNSW-NB 15 | Univariate and principal component analysis (PCA) | ANN, SVM | 91%, 92% |
Meftah et al. (2019) | UNSW-NB 15 | Random forest | GBM, LR, SVM | 61.83%, 77.21%, 82.11% |
Injadat et al. (2020) | CICIDS 2017, UNSW-NB 15 | Correlation based (CBFS) and information gain (IGBFS) | KNN RF | 99% for both datasets |
Belgrana et al. (2021) | NSL-KDD | Condensed nearest neighbors (CNN) | Radial basis function (RBF), CNN | 94.28%, 95.54% |
Lee et al. (2020) | CICIDS2017 dataset | Autoencoder | Deep sparse autoencoder random forest (DSAE-RF) | 99.83% |
Gu and Lu (2021) | UNSW-NB 15, CICIDS2017, NSL-KDD, and Kyoto 2006 + | Feature transformation with Naïve Bayes | SVM | 98.92% (CICIDS2017), 99.35% (NSL-KDD), 93.75% (UNSW NB15), 98.58% (Kyoto 2006 +) |
Moustafa (2021) | Network TON_IoT | Wrapper feature selection technique-based RF | GBM RF DNN | 93.83%, 99.98%, 99.92% |