Open Access
Table 4.
Summary of comparison
| Metric | Proposed Cso-Xg-Boost | Baseline Xg-Boost | Other Ids solutions with tuning |
|---|---|---|---|
| Performance | High accuracy, precision, F1-score | Moderate without tuning | Varies, often lower or comparable |
| Optimization Efficiency | Efficient, balances exploration/exploitation | No optimization, less efficient | Slower (Grid Search) or less thorough (Random Search) |
| Generalization | Robust, reduces overfitting | Moderate, possible overfitting | Depends on model, usually lower |
| Handling Imbalanced Data | Excellent, optimized parameters | Moderate, default parameters | Depends on model, often requires resampling |
| Computational Complexity | Low to moderate, scalable | Efficient but not optimized | Varies, typically higher than XG-Boost |
| Flexibility | High, adaptable to various tasks | Limited without tuning | Moderate, depends on the solution |
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.