Prediction of Road Accidents Using Machine Learning Algorithms

The rapid development of AI shows its power and great development potential in practical engineering applications. Critical issues and potential solutions can reduce road traffic accidents and application of AI in road accident prediction. The published use of AI for road accident prediction is reviewed, presented, and represented as the main objective. The methods are collecting article data, quotations, presentation, and representation. The article data collection was obtained from 671 conference and journal articles in 2019-2023, but the suitability of articles that can be used is 69. Quotation produces a grouping of approaches, models, predictions, and benefits. The presentation showed that most approaches used were machine learning, the most used model was random forest, the prediction was mostly about severity, and the most benefit was about number reduction. Representation produces road accidents and related factors into factors in road accident predictions using artificial intelligence, so strategies and anticipation can be made to overcome them to improve road safety. AI in road accident prediction plays an important role in building predictive models with the hope that road accidents can be identified early, risk factors can be reduced, and effective preventive measures can be taken to improve road safety.

A study on road accident prediction and contributing factors using explainable machine learning models: analysis and performance

Due to population growth and the increasing number of vehicles on rural roads, traffic accidents have become one of the most important problems in the transportation system, which greatly affects the social and economic situation of the people. The main purpose of this study was to apply the analytical method to investigate the factors affecting the severity of traffic accidents on rural roads of Guilan, Iran, in order to determine the most effective factor in the occurrence of these accidents. At first, the frequency analysis was used to evaluate the variables and their frequency, then the Friedman test (FT) was applied to prioritize the factors, and the exploratory factor analysis (EFA) was used to determine the most effective factor in the occurrence of vehicle accidents in Guilan rural roads. Based on the FT, weather condition was the most important factor effective in these accidents. According to the results of the EFA, five factors were identified as the main factors involved in accidents in which the first factor contributing to accidents was the environmental factor, including weather condition and road surface condition. This indicates that concurrent result of the FT and the EFA, weather condition as an environmental factor, was identified as the most important factor affecting vehicle accidents on rural roads of Guilan. Finally, safety strategies were proposed to increase safety and reduce accidents along these roads.

The rapid spread of contagious diseases poses a colossal threat to human existence. Presently, the emergence of coronavirus COVID-19 which has rightly been declared a global pandemic resulting in so many deaths, confusion as well as huge economic losses is a challenge. It has been suggested by the World Health Organization (WHO) in conjunction with different Government authorities of the world and non-governmental organizations, that efforts to curtail the COVID-19 pandemic should rely principally on measures such as social distancing, identification of infected persons, tracing of possible contacts as well as effective isolation of such person(s) for subsequent medical treatment. The aim of this study is to provide a framework for monitoring Movements of Pandemic Disease Patients and predicting their next geographical locations given the recent trend of infected COVID-19 patients absconding from isolation centres as evidenced in the Nigerian case. The methodology for this study, proposes a system architecture incorporating GPS (Global Positioning System) and Assisted-GPS technologies for monitoring the geographical movements of COVID-19 patients and recording of their movement Trajectory Datasets on the assumption that they are assigned with GPS-enabled devices such as smartphones. Accordingly, fifteen (15) participants (patients) were selected for this study based on the criteria of residency and business activity location. The ensuing participants movements generated 157, 218 Trajectory datasets during a period of 3 weeks. With this dataset, mining of the movement trace, Stay Points (hot spots), relationships, and the prediction of the next probable geographical location of a COVID-19 patient was realized by the application of Artificial Intelligence (AI) and Data Mining techniques such as supervised Machine Learning (ML) algorithms (i.e., Multiple Linear Regression (MLR), k-Nearest Neighbor (kNN), Decision Tree Regression (DTR), Random Forest Regression (RFR), Gradient Boosting Regression (GBR), and eXtreme Gradient Boosting regression(XGBR) as well as density-based clustering methods (i.e., DBSCAN) for the computation of Stay Points (hot spots) of COVID-19 patient. The result of this study showed clearly that it is possible to determine the Stay Points (hot spots) of a COVID-19 patient. In addition, this study demonstrated the possibility of predicting the next probable geographical location of a COVID-19 patient. Correspondingly, Six Machine Learning models (i.e., MLR, kNN, DTR, RFR, GBR, and XGBR) were compared for efficiency, in determining the next probable location of a COVID-19 patient. The result showed that the DTR model performed better compared to other models (i.e., MLR, kNN, RFR, GBR, XGBR) based on four evaluation matrices (i.e., ACCURACY, MAE, MSE, and R2) used. It is recommended that less developed Countries consider adopting this framework as a policy initiative for implementation at this burgeoning phase of COVID-19 infection and beyond. The same applies to the developed Countries. There is indication that GPS Trajectory dataset and Machine Learning algorithms as applied in this paper, appear to possess the potential of performing optimally in a real-life situation of monitoring a COVID-19 patient. This paper is unique given its ability to predict the next probable location of a COVID-19 patient. In the review of extant literature, prediction of the next probable location of a COVID-19 patient was not in evidence using the same Machine Learning algorithms.

Objectives: To observe the road traffic accidents pattern across the states in India and to conclude whether the expenditure on road and transportation infrastructure is really justified with the traffic safety across the country. To decide the road safety policies across the states in India using backward elimination approach of regression model. Methods/Analysis: This study attempts to show statistical relations between road traffic accidents with predictive variables; number of vehicles registered, total length of roads, total number of road accidents and total expenditure on transportation infrastructure in the states of India. Time series data covering the period 2001 to 2013 was used on yearly road traffic accidents and other predictive variables for all 28 states of India. Findings: In recent decades, road accidents are becoming one of the most growing concerns in India. As mentioned in report of Ministry of Road Transport & Highways Government of India, road accidents impose significant costs which are about 3% GDP for India since 1999 to 2000. In 2004, India was 9th leading cause of death and expected to be 5th leading cause of death by 2030 worldwide. These facts are enough to catch attention of Government agencies and researchers towards safer traffic solutions. The results from the analysis shows few findings like expenditure on transportation infrastructure may not be an important predictive variable for road traffic accidents for pan India situations. Finally, if all the states of India partitioned in different clusters according to the behavior of states, the study states that there exists a strong positive correlation between road traffic accidents, number of vehicles registered and total length of roads. The relationship found to be statistically significant at 5% level of significance for each cluster of states. Applications/Improvements: The results obtained in study can assist transportation departments and Government bodies to implement better strategies across the country to ensure road traffic safety and reduce losses in terms of money and infrastructure costs.Keywords: Policies in India, Regression Analysis, Road Accidents, Transportation, Road Safety

Road accidents are one of the major causes of deadly injuries and deaths. It is possible to predict the possibility of accidents by studying past data. The occurrence of road accidents is associated with multiple factors such as speed, traffic condition, day, time, weather conditions, road construction, etc. Machine learning Algorithms are used to achieve the goal. The key steps involved are data pre-processing, Training the model with supervised learning concepts and creation of the interactive Dashboard. The supervised learning algorithms like Random Forest and Logistic regression are tried and assessed on the basis of accuracy and performance for the training of the model, along with the DBSCAN algorithm for the clustering of the data. The outcome of this project will benefit the public in providing a visualization tool that will evaluate the probability of an accident. In addition, it will help the traffic department in implementing strategies to reduce road accidents.

For a better driving experience and fewer traffic accidents, road surface condition is critical. Traditional road condition monitoring systems lack the temporal (speed) and geographical (coverage) responses required for overall road quality maintenance. Several alternative methods that use sensors installed on cars have been presented. To ensure road safety, the safe system approach focuses on designing safe cars, road networks, and road users. As a result of the World Health Organization (WHO), this technique is becoming increasingly popular across the world. In this research study, a short-to-medium term dynamic evaluation of road safety is proposed. Here, we provide an innovative, cost-effective Internet of things (IoT) architecture that allows the creation of a robust and dynamic computing core for evaluating the safety of road networks and their elements. The machine learning technique described here may be used to monitor roads for problems that pose a safety concern to commuters and to give maintenance information to appropriate authorities on a bigger scale.KeywordsIoTWHORoad safetyRoad networkDynamic

A review on neural network techniques for the prediction of road traffic accident severity

With the rise of Advanced Driver Assistant Systems (ADAS) and the introduction of Highly Automated Driving (HAD), understanding and predicting road traffic accidents becomes increasingly important. Especially for the assessment of HAD/ADAS systems and of road safety, the precise prediction of the system’s impact on the occurrence of road traffic accidents is essential.Traffic simulations, as one option of virtual assessment, enable the assessment of safety systems in virtual test fields. By modelling the human driver, it is possible to simulate and predict future accident constellations and accident severities. In addition, the influence of vehicle factors, such as the presence of safety systems or their characteristics, on the occurrence of accidents can be investigated.This article elucidate virtual assessment by executing the whole procedure for an automatic emergency braking (AEB) system capable to detect cyclists at urban intersections. The authors implement a new driver behaviour model (DReaM) to predict road traffic accidents at urban intersections. Using the AEB as an example, the influence of different sensor opening angles on crash frequency is investigated by varying the sensors in three steps: 100°, 180°, and 210°. A total of 240,000 situations are simulated of which 6,674 are crashes. Within the two investigation scenarios, the crash frequency is reduced by up to 88.43%, or 93.92% by introducing the AEB system.The article shows that virtual assessment enables the prediction of new safety systems regarding road safety at early design stages. In addition, different system characteristics can be compared efficiently, e.g., the sensor opening angle.

Road traffic accidents (RTAs) in Northwest Ethiopia, a region with a fatality rate of 32.2 per 100,000 residents, pose a critical public health challenge exacerbated by infrastructural deficits and environmental hazards. This study leverages machine learning (ML) to predict accident severity, addressing gaps in localized predictive frameworks for low- and middle-income countries (LMICs). Our study aims to predict the severity of car accidents in Northwest Ethiopia via machine-learning techniques. Using a dataset of 2,000 accidents (2018–2023) from police reports, we integrated driver demographics, behavioral factors (e.g., alcohol use, seatbelt compliance), and environmental conditions (e.g., unpaved roads, weather) in North West Ethiopia. Ten ML models, including Random Forest, XGBoost, and LightGBM, were evaluated after addressing class imbalance via the Synthetic Minority Oversampling Technique (SMOTE). Hyperparameter tuning and Shapley Additive explanations (SHAP) provided model optimization and interpretability. Random Forest outperformed other models, achieving 82% accuracy (AUC-ROC: 0.87) post-tuning. Driver age (mean: 44 years) and environmental factors (e.g., nighttime on unlit roads, rainy conditions) were critical predictors, increasing fatal accident likelihood by 62%. SMOTE improved the accuracy of the outperforming random forest accuracy from 78.6 to 82%. Random Forest exhibited the highest recall (0.82) after optimization, while ensemble methods dominated performance metrics. The study underscores the efficacy of ML in contextualizing accident severity in LMICs, with Random Forest emerging as a robust tool for policymakers. Prioritizing road paving, sobriety checkpoints, and motorcycle safety could mitigate risks, aligning with Sustainable Development Goal 3.6. Future work should address data limitations (underreporting, geospatial gaps) and expand model interpretability.

Aim: This study intends to assess drivers’ state of knowledge on the main causes and the best strategies capable of reducing road traffic accidents in the Bongo District of Ghana.
 Study Design: The study was a descriptive design.
 Place and Duration of Study: The study was conducted in the Bongo District of the Upper East Region of Ghana between July 2023 and August 2023.
 Methodology: The researcher conducted a survey involving 100 drivers who were randomly selected from a driver population of 120 in the Bongo District. The data were analyzed using Microsoft Excel [Version 19]. The results of the data analysis were presented as tables and charts.
 Results: The study found that the main causes of road traffic accidents emanate from human error such as low driving skills, excessive speeding, and poor vision of driver. Also, road conditions like bad road networks, no lane markings, and no speed limit signs were found as causal factors of road traffic accidents. Faulty vehicles with causative factors such as defective tyres, brakes, and lights were noted to be also responsible for road accidents. Nevertheless, the least factor was environmental conditions such as bad weather conditions and stray animals being enumerable as environmental conditions which cause road accidents. The study further found that the best strategies for curbing road traffic accidents were enforcement of traffic laws, education and training of drivers, obeying traffic rules by drivers, and maintaining vehicles well.
 Conclusion: The study, therefore, concluded that human factors play a central role in contributing to road accidents and curbing road traffic accidents. The research findings can be used to guide stakeholders in formulating and prioritizing interventions that are more human-centred for the curbing of road traffic accidents in the Bongo District of Ghana.

ObjectiveEnhanced technology in computer and internet has driven scale and quality of data to be improved in various areas including healthcare sectors. Machine Learning (ML) has played a pivotal role in efficiently analyzing those big data, but a general misunderstanding of ML algorithms still exists in applying them (e.g., ML techniques can settle a problem of small sample size, or deep learning is the ML algorithm). This paper reviewed the research of diagnosing mental illness using ML algorithm and suggests how ML techniques can be employed and worked in practice.MethodsResearches about mental illness diagnostic using ML techniques were carefully reviewed. Five traditional ML algorithms-Support Vector Machines (SVM), Gradient Boosting Machine (GBM), Random Forest, Naïve Bayes, and K-Nearest Neighborhood (KNN)-frequently used for mental health area researches were systematically organized and summarized.ResultsBased on literature review, it turned out that Support Vector Machines (SVM), Gradient Boosting Machine (GBM), Random Forest, Naïve Bayes, and K-Nearest Neighborhood (KNN) were frequently employed in mental health area, but many researchers did not clarify the reason for using their ML algorithm though every ML algorithm has its own advantages. In addition, there were several studies to apply ML algorithms without fully understanding the data characteristics.ConclusionResearchers using ML algorithms should be aware of the properties of their ML algorithms and the limitation of the results they obtained under restricted data conditions. This paper provides useful information of the properties and limitation of each ML algorithm in the practice of mental health.

The recent advance in information technology has created a new era named the Internet of Things (IoT). This new technology allows objects (things) to be connected to the Internet, such as smart TVs, printers, cameras, smartphones, smartwatches, etc. This trend provides new services and applications for many users and enhances their lifestyle. The rapid growth of the IoT makes the incorporation and connection of several devices a predominant procedure. Although there are many advantages of IoT devices, there are different challenges that come as network anomalies. In this research, the current studies in the use of deep learning (DL) in DDoS intrusion detection have been presented. This research aims to implement different Machine Learning (ML) algorithms in WEKA tools to analyze the detection performance for DDoS attacks using the most recent CICDDoS2019 datasets. CICDDoS2019 was found to be the model with best results. This research has used six different types of ML algorithms which are K_Nearest_Neighbors (K-NN), super vector machine (SVM), naïve bayes (NB), decision tree (DT), random forest (RF) and logistic regression (LR). The best accuracy result in the presented evaluation was achieved when utilizing the Decision Tree (DT) and Random Forest (RF) algorithms, 99% and 99%, respectively. However, the DT is better than RF because it has a shorter computation time, 4.53 s and 84.2 s, respectively. Finally, open issues for further research in future work are presented.

Road accidents are still a prominent urban issue, worsened by a growing population and uncertain weather patterns. Conventional accident prediction models use static historical data, which does not enable them to be responsive to real-time traffic patterns. This research is responding to the demand for predictive models that are adaptive, smart, and can facilitate smart city infrastructure through real-time data integration. Machine learning models—Multilayer Perceptron, Gradient Boosting, Random Forest, Support Vector Machine, and K-Nearest Neighbors—were tested with both historical and real-time traffic data. The models were optimized and trained on varied datasets to improve prediction accuracy. Of these, Gradient Boosting recorded the highest accuracy at 88.1%, followed by Random Forest at 83.73%, showing the power of ensemble learning techniques in predicting accidents. This research emphasizes the significance of real-time data integration for accident prediction and prevention. Merging environmental elements like weather conditions and traffic congestion improves prediction quality, allowing proactive prevention of accidents. By leveraging these models, cities can shift towards data-based road management, infrastructure planning, and congestion control. The results show that real-time models of accident prediction have the potential to enhance urban safety, opening the door to smarter and more efficient traffic management systems.

Road traffic accidents are the world’s number one cause of death among young people between 16 and 24. Bangladesh has one of the highest fatality rate in road accidents in the world. Moreover, Dhaka, the capital of Bangladesh, is the most vulnerable city both in terms of total number of accidents and accident rates. A total number of 2,720 accidents occurred within 2007-2011. This has caused a total of 1,481 numbers of pedestrian fatal accidents with 1,562 pedestrian fatal casualties. At this backdrop, road safety problem has become one of the major issues for the transport regulators and traffic law enforcers in Dhaka. The objective of this research is to identify the most accident-prone roads, the victims as well the in-depth reasons for traffic accident in Dhaka Metropolitan Area. In this article, the road safety situation and traffic accident trends are discussed elaborately. It also discusses some contemporary issues and priorities in addressing the road safety problems. Based on the findings of this study, some recommendations related to Engineering, Enforcement, Educational and Emergency Response aspects are made. This kind of research will help generating new ideas to find out the parameters for reducing road traffic accidents in greater Dhaka City. Vol. 6, December 2013, pp. 103-118