Using Hybrid Deep Learning Approach to Enhanced Network Intrusion Detection with Spatial-Temporal Feature Integration
Bibliographic record
Abstract
Intrusion Detection Systems (IDS) play a vital role in network security by detecting and preventing malicious activities.The network intrusion data is integrated into a vast number of common occurrences due to the dynamic and ever-changing networking environment.This results in a scarcity of training cases for models and detection outcomes, accompanied by a significant percentage of false detections.Our suggested Network-IDS addresses the issue of data imbalance by integrating Deep Learning Networks (DLN) via hybrid sampling.We begin by collecting out-of-the-ordinary samples from the majority and eliminating them using the Difficult-Set-Sampling-Technique method, which stands for Difficult-Set-Sampling-Technique (DSST).Next step is to increase the minority group's sample size using (DCGAN) means Deep-Convolutional-Generative-Adversarial-Networks.Step two involves building a model for a deep neural network to extract geographical features using DenseNet169, in addition, we utilize SAT-Net to capture features of temporal.This approach effectively represents the unique attributes of the dataset.Lastly, we deployed the EESNN to identify assault types.In addition to that, we conducted tests on the latest and most extensive intrusion datasets, the Telecommunications Network Internet of Things (ToN-IoT) dataset as well as the CICIDS2019 dataset, to verify of proposed approach.The outcome demonstrates that our recommended structure surpasses similar efforts in terms of accuracy, false alarm rate, recall, and precision.The findings indicate that our proposed system is superior to other attempts of a similar kind in terms of accuracy, false alarm rate, recall, and precision.We will provide a detailed explanation of this in the comparative section.
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 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 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 | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
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".