Efficient Traffic Classification Using Hybrid Deep Learning
Bibliographic record
Abstract
Network traffic classification provides an essential contribution in network administration functions and network management such as QoS, security, and billing. Those functions need a timely and accurate detection of specific traffics. Current network traffic classification methods offer supervised and unsupervised learning capabilities for network traffic prediction or classification. Classical machine learning classifiers that use a single classification model suffer from low prediction and classification accuracy, especially for high dimensional datasets with a high sparsity level. These challenges in individual-based learning models have created a need for hybrid learning. Recently, hybriddeep learning has shown a significant role in traffic forecasting and classification due to its efficiency. However, a tradeoff between the aggregate models and the classification accuracy presents a substantial challenge in network traffic classification problems. In this paper, we have suggested two hybrid models that combine the Convolutional Neural Network (CNN) along with the Recurrent Neural Network (RNN) models, inclusive of the Gated recurrent unit (GRU) and Long Short-Term Memory (LSTM), to improve traffic classification accuracy. The efficiency of the suggested models has been evaluated by comparing them with various individual-based models using real network traffic traces. The hybrid CNN-LSTM and CNN-GRU have achieved an accuracy of up to 99.23% and 93.92%, respectively, for binary classification and 67.16% for multiclass classification.
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) | 0.001 | 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 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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".