Enhancing Iot Intrusion Detection With Transformer-Based Network Traffic Classification
Bibliographic record
Abstract
The fast rise of cyber threats has exposed flaws in traditional intrusion detection systems (IDS), particularly in the Internet of Things (IoT) context. Large language models (LLMs) and other deep learning techniques provide sophisticated skills for recognizing intricate attack patterns, which can enhance IDS capabilities. Using two benchmark datasets, we conducted a comparative analysis of IoT network intrusion detection using encoder-based Transformer models and two baseline deep learning models. Experiments on the NF-BoT-IoT and NF-ToN-IoT datasets showed that the evaluated transformer models (BERT-tiny, Electrasmall, and DistilBERT) performed well in binary and multi-class classification tasks. By utilizing self-attention mechanisms, these models surpass the baseline models in capturing crucial context. According to our experiments, the best binary classification accuracies were 99.38 % and 99.98 % for the NF-BoT-IoT and NF-ToN-IoT datasets, respectively. In multi-class classification, DistilBERT achieved the highest accuracy (84.35 %) on the NF-BoT-IoT dataset, whereas BERT-tiny achieved the highest accuracy (<tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{7 2. 7 \%}$</tex>) on the NF-ToN-IoT dataset. For the Baseline models, multi-layer perceptron (MLP) demonstrated competitive binary classification performance with low processing costs, whereas long short term memory (LSTM) struggled with multiclass classification, highlighting Transformer models' superiority for high-dimensional, multi-class data. Our findings validate Transformer models' suitability for sophisticated intrusion detection systems, providing significant improvements in detection accuracy and overall efficiency across complicated attack scenarios in IoT networks.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.002 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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 teacher head, 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".