A Predictive Framework of Speed Camera Locations for Road Safety
Bibliographic record
Abstract
Road traffic crashes are a public health issue due to their terrible impact on individuals, communities, and countries. Studies affirmed that vehicle speed is a major contributor to crash likelihood and severity. At the same time, they identified Automated Speed Enforcement (ASE) systems, namely speed cameras, as a highly effective measure to reduce excessive and inappropriate speed, and thus improving road safety. However, identifying optimum sites for fixed speed camera placement stays an open issue in the literature, although it is a key factor that guarantees the efficiency of such ASE systems. This paper describes a predictive framework of speed camera locations using a classification algorithm that can predict, for each section of a given road network, its pertinence as a speed camera location. First, we identify a set of features as predictors of the classification algorithm, that we have argued their goodness through correlation tests. Second, for training our algorithm, data from road controlled sections, corresponding to existing speed cameras, is exploited. Each section class reflects the contribution level of the ASE system (good, neutral, or bad) to road safety. Third, as a proofof-concept, the framework has been implemented and deployed on the Moroccan road network. The results showed that Random Forest classifier is the best performing model attaining an accuracy of 95% and a precision of 88%. Further, a tool was developed to visualize updated classification results on a Moroccan road network map to support authorities in their decision making process.
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 | 0.000 |
| 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".