A Comparative Study of Machine Learning Algorithms to Predict Road Accident Severity
Bibliographic record
Abstract
Road accidents is a global issue that cause deaths and injuries besides other direct and indirect losses. Countries and international organisations have designed technologies, systems, and policies to prevent accidents. The use of big traffic data and artificial intelligence may help develop a promising solution to predict or reduce the risk of road accidents. Most existing studies examine the impact of road geometry, environment, and weather parameters on road accidents. However, human factors such as alcohol, drug, age, and gender are often ignored when determining accident severity. In this work, we considered various contributing factors and their impact on the prediction of the severity of accidents. For this, we studied a set of single and ensemble mode machine learning (ML) methods and compared their performance in terms of prediction accuracy, precision, recall, F1 score, area under the receiver operator characteristic (AUROC). This research considered the road accident severity prediction as a classification problem that can classify the intensity of an accident in two categories: (i) binary classification (e.g. grievous and non-grievous), and (ii) multiclass classification (fatal, serious, minor, and non-injury). Our results show that Random Forest (RF) outperformed other methods' like logistic regression (LR), K-nearest neighbor (KNN), naive Bayes (NB), extreme gradient boosting (XGBoost), and adaptive boosting (AdaBoost) (e.g., 86.64% for binary and 67.67% for multiclass classification) in both single and ensemble ML methods in comparison to other methods considered in this research. LR, KNN, and NB are single mode ML methods that show similar performance to each other for both binary and multiclass classification. Compared to single mode, ensemble ML methods can predict the severity of accident more accurately with the order of RF, XGBoost, and Adaboost. The findings from this study can help to gain insights into the accident contributing factors and the severity of injuries as a result.
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 | 0.012 | 0.022 |
| Meta-epidemiology (narrow) | 0.002 | 0.000 |
| Meta-epidemiology (broad) | 0.002 | 0.002 |
| Bibliometrics | 0.007 | 0.004 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.002 | 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".