Enhancing Network Security: A Study on Classification Models for Intrusion Detection Systems
Bibliographic record
Abstract
Computer users face a constant influx of internet packets, ranging from legitimate ones to those sent by malicious entities. With the exponential growth in user numbers and evolving attack types, traditional countermeasure methods are becoming ineffective. Artificial intelligence (AI) techniques offer a promising solution to address these challenges. This study leverages AI methods to develop nine classification models using supervised machine learning classifiers. The author has implemented several machine learning models, including bagging, multi-layer perceptron, logistic regression, extreme gradient boosting, and random forest. The authors utilize three datasets (Knowledge Discovery in Databases 1999 dataset, used for network intrusion detection research), UNSW-NB15 (a dataset capturing contemporary network attack patterns generated at the University of New South Wales), and CICIDS2017 (Canadian Institute for Cybersecurity Intrusion Detection System dataset, containing modern attack scenarios)(KDD99, UNSW NB15, and CICIDS2017) with varying train-test ratios to train the classifiers. The author employs accuracy and F1 score metrics to evaluate the model’s performance. The Extreme Gradient Boosting classifier exhibits the highest performance across all three datasets, especially with an 80% feature reduction. Various oversampling and undersampling techniques balance the dataset to improve falsenegative rates. Performance metrics show improvements across all dataset types, with extreme gradients boosting accuracy. The meta-ensemble learning model does better at sub-multiclass classification than decision trees, random forests, and extreme gradient boosting. It also does better than logistic regression and multi-layer perceptron in multiclass classification. Two hidden layers achieved the highest accuracy for binary classification on the KDD99 dataset. Multiclass classification presents challenges with identifying minor classes, but performance improves with additional hidden layers. Random Forest outperforms other classifiers in accuracy, which is consistent with simulation results.
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 | 0.006 | 0.020 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.003 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.003 | 0.006 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.002 | 0.003 |
| 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".