Protection of Critical Emergency Response Infrastructures through Machine Learning
Bibliographic record
Abstract
Critical Emergency Response Infrastructures are essential to supporting the goals of sustainable development, especially during crises or emergencies that disrupt sustainable progress. While they provide the framework that enables emergency response operations to be effective and resilient, Public Safety Answering Points (PSAP) are the operational frontline of emergency response. A PSAP is the communications center where emergency calls made by the public are received (e.g., by dialing 911 in the United States and Canada, 112 in Europe, or equivalent local emergency numbers).The increasing reliance on PSAPs to ensure the appropriate response of civil protection services requires the implementation of advanced mechanisms to protect these essential infrastructures against possible unauthorized increases in call volume, which have the potential to affect citizen assistance in critical situations. This study employs statistical and machine learning techniques to analyse call patterns in PSAPs to identify and mitigate the effect of potentially malicious non-emergency calls. By using data analysis methods and pattern recognition techniques, it is feasible to characterise genuine emergency calls and differentiate them from those made with the intention of overloading the system, thus ensuring that response times for real emergencies are not compromised. Starting with the generation of a synthetic dataset that emulates the most significant features of a Telephony Denial of Service, along with an extensive set of real data, this research follows a multidimensional approach by combining temporal-spatial analysis and machine learning algorithms to characterize the behavioral patterns of emergency calls, using Gaussian Mixture Model. The obtained results identify the Gaussian Mixture Model (GMM), a probabilistic model that assumes that data points are generated from a mixture of a finite number of Gaussian distributions with unknown parameters, as an optimal solution for this characterization.
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 | 0.003 |
| 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".