Feature Selection for Robust Spoofing Detection: A Chi-Square-based Machine Learning Approach
Bibliographic record
Abstract
Nowadays, the Internet of Things (IoT) system is vulnerable to spoofing attacks that can easily where attackers can easily pose as a legal entity of the network. A “spoofing attack” refers to a type of cyber-attack when an attacker purposefully impersonates or masquerades as someone or something else to deceive the target or obtain unauthorized access to systems, information, or resources. In such attacks, the attacker alters their name, IP address, or other attributes to fool the victim into thinking they are engaging with a legitimate entity. Spoofing attacks can take place via a variety of channels, including, ARP and DNS spoofing. Therefore. Spoofing attacks can have serious consequences. We proposed a new approach based on three machine learning models LightGBM, Gradient Boost, and XGBoost to classify attacks on spoofing, we used Chi-square to select the best features to get the highest performance, and we demonstrated that the results using Chi-square achieved higher results than without Chi-square and improved the result with a rate three percent of accuracy. In terms of results, LightGBM outperformed other models by achieving 89%, 91 %, 87 %, and 89 % for accuracy, precision, recall, and f1-score, respectively. This shows the potential and efficiency of Chi-square to achieve the best performance by selecting the best features, thus providing a secure system and identifying cyber-attacks on systems.
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 | 0.004 | 0.007 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.003 | 0.002 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.001 | 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".