Table 1 Summary of the existing studies on IDS developed for various networks.
Year (Reference) | Dataset | Algorithm(s) | Limitation(s) | Accuracy |
|---|---|---|---|---|
201832 | NSL-KDD | Artificial-Bee Colony and Fuzzy C-Means Clustering | • Limited scalability. • Slow convergence | 99.00% |
201940 | NSL-KDD | Beep belief networks (DBN) and improved genetic algorithm (IGA) | • Struggle with feature selection. • Slow convergence | 99.8% |
202041 | ISCX 2012 | Adaptive-Grasshopper Optimization Algorithm and SVM | • Limited scalability. • Slow convergence | 99.13% |
202154 | UNSW-NB15 | Tabu Search and Random Forest | • Low accuracy • Reduced response time | 83.12% |
202133 | NSL-KDD | Random forest, K-nearest neighbour and XGBoost | • Limited scalability. | 99.34% |
202142 | UNSW-NB15 | Particle Swarm Optimization | • Low accuracy • Reduced response time | 86.68% |
202143 | NSL-KDD | Artificial Neural Networks and SVM | • Limited scalability. | 90.45% |
202129 | TON_IoT | Convolutional Neural Network and Recurrent Neural Network | • Reduced response time | 99.92% |
202244 | NSL-KDD | Butterfly Optimization Algorithm (BOA) and ANN | • Limited scalability. | 93.27% |
202230 | Bot-IoT | ANN | • Limited scalability. | 99.611% |
202345 | CICIoT2023 | Random Forest and Deep Neural Network | • Low accuracy • Reduced response time | Average accuracy is below 80% |
202346 | ToN-IoT | RepuTE algorithm | • Limited Scalability • Reduced response time | 99.90% |
202347 | CIC IDS2017, N-BaIoT dataset and NF-ToN-IoT | Deep Learning based Bi-LSTM technique | • Limited Scalability | 99.67% |
202451 | NF-ToN-IoT | A hybrid deep learning technique utilizing LSTM and one-dimensional CNN | • Low accuracy • Reduced response time | 98.75% (M) |
