Advanced NID-VGG16 with Orca Predation Optimization Based 1DCNN-BiLSTM for Network Intrusion Detection
Bibliographic record
Abstract
Many cybersecurity dangers exist for modern computer networks, which emphasize how crucial Network Intrusion Detection (NID) is necessary. The ever-changing threat landscape has made it difficult for traditional intrusion detection systems (IDS) to remain successful, which emphasizes the growing significance of artificial intelligence (AI) solutions. To improve threat detection effectiveness, the paper used Deep Learning-powered network intrusion detection (DL-powered NID) in this study. The tests utilized two datasets: the Canadian Institute for Cybersecurity 2017 (CICIDS2017) dataset and Internet of Things network traffic (IOT23) captures to verify the efficacy of this technique. Preprocessing the dataset entails applying the Preprocessing and Minimax Scaling (PMS) method, which includes filtering, transforming, and normalizing the data. The paper offer the NID-VGGI6 framework, a 16-convolution-layer network intrusion detection visual geometry group based on Convolutional Neural Networks (CNNs), for feature extraction. In order to provide a class and scale-invariant architecture, this framework combines multilevel and multiscale features with data augmentation approaches. During NID-VGGI6 training, the focal loss function solves class imbalance, and the Flatten-T Swish (FTS) activation function reduces gradient vanishing and explosion problems. In order to improve decision-making in NID, a one-dimensional CNN-based Bidirectional Long Short-Term Memory (1DCNN-BiLSTM) model is used for classification after feature extraction. The paper utilized the Orca Prediction Optimization Algorithm (OPOA) to fine-tune classification hyperparameters for maximum accuracy. The findings show that the suggested model performs better than the existing ones and reaches an impressive 99.9% accuracy rate for both the datasets.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.002 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.005 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".