Table 1.
Comparative analysis of selected IDS studies.
Fig 1.
Proposed system architecture.
Table 2.
Class distribution before merging.
Table 3.
Attack category distribution (pre-binarization).
Table 4.
Cross-dataset feature harmonization.
Table 5.
Hyperparameter configuration and search space for evaluated models.
Table 6.
Class distribution of cybersecurity datasets.
Table 7.
Class-wise metrics and confusion matrix (random forest, merged dataset).
Table 8.
Weighted average metrics for top models (merged dataset).
Table 9.
Training time and efficiency trade-offs.
Table 10.
Class distribution in merged dataset (before and after SMOTE).
Fig 2.
Training vs validation accuracy across folds.
Table 11.
Cross-dataset generalization.
Table 12.
Cross-validation results on individual datasets.
Table 13.
Cross-dataset performance drop.
Table 14.
Confusion matrix (aggregated) – Random forest (on harmonized merged dataset).
Fig 3.
Cross-dataset ROC curves.
Fig 4.
Performance comparison bar chart (within vs cross dataset).
Table 15.
Ablation study (merged dataset, random forest).
Fig 5.
Performance impact of each pipeline component.
Table 16.
Tamper-evident hash log structure.
Fig 6.
Tamper-evident blockchain-style prediction log (cryptographic hash trace).
Table 17.
Statistical significance test results.
Fig 7.
Statistical significance test (Mc Nemer/Wilcoxon).
Table 18.
Average performance across all settings (NSL_ONLY, CIC_ONLY, MERGED).
Table 19.
Evaluation metrics reporting summary.
Table 20.
Performance metrics and efficiency trade-offs.
Table 21.
Recent State-of-the-Art Intrusion Detection System (IDS) approaches.