Dataset Information

Predicting Risky and Aggressive Driving Behavior among Taxi Drivers: Do Spatio-Temporal Attributes Matter?

ABSTRACT: Risky and aggressive driving maneuvers are considered a significant indicator for traffic accident occurrence as well as they aggravate their severity. Traffic violations caused by such uncivilized driving behavior is a global issue. Studies in existing literature have used statistical analysis methods to explore key contributing factors toward aggressive driving and traffic violations. However, such methods are unable to capture latent correlations among predictor variables, and they also suffer from low prediction accuracies. This study aimed to comprehensively investigate different traffic violations using spatial analysis and machine learning methods in the city of Luzhou, China. Violations committed by taxi drivers are the focus of the current study since they constitute a significant proportion of total violations reported in the city. Georeferenced violation data for the year 2016 was obtained from the traffic police department. Detailed descriptive analysis is presented to summarize key statistics about various violation types. Results revealed that over-speeding was the most prevalent violation type observed in the study area. Frequency-based nearest neighborhood cluster methods in Arc map Geographic Information System (GIS) were used to develop hotspot maps for different violation types that are vital for prioritizing and conducting treatment alternatives efficiently. Finally, different machine learning (ML) methods, including decision tree, AdaBoost with a base estimator decision tree, and stack model, were employed to predict and classify each violation type. The proposed methods were compared based on different evaluation metrics like accuracy, F-1 measure, specificity, and log loss. Prediction results demonstrated the adequacy and robustness of proposed machine learning (ML) methods. However, a detailed comparative analysis showed that the stack model outperformed other models in terms of proposed evaluation metrics.

SUBMITTER: Zahid M

PROVIDER: S-EPMC7312618 | biostudies-literature | 2020 Jun

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Predicting Risky and Aggressive Driving Behavior among Taxi Drivers: Do Spatio-Temporal Attributes Matter?

Zahid Muhammad M Chen Yangzhou Y Khan Sikandar S Jamal Arshad A Ijaz Muhammad M Ahmed Tufail T

International journal of environmental research and public health 20200602 11

Risky and aggressive driving maneuvers are considered a significant indicator for traffic accident occurrence as well as they aggravate their severity. Traffic violations caused by such uncivilized driving behavior is a global issue. Studies in existing literature have used statistical analysis methods to explore key contributing factors toward aggressive driving and traffic violations. However, such methods are unable to capture latent correlations among predictor variables, and they also suffe ...[more]

PMID: 32498347

Similar Datasets

Project description:ObjectiveTwo aims of this study were to develop and validate A) a metric to identify drivers with type 1 diabetes at high risk of future driving mishaps and B) an online intervention to reduce mishaps among high-risk drivers.Research design and methodsTo achieve aim A, in study 1, 371 drivers with type 1 diabetes from three U.S. regions completed a series of established questionnaires about diabetes and driving. They recorded their driving mishaps over the next 12 months. Questionnaire items that uniquely discriminated drivers who did and did not have subsequent driving mishaps were assembled into the Risk Assessment of Diabetic Drivers (RADD) scale. In study 2, 1,737 drivers with type 1 diabetes from all 50 states completed the RADD online. Among these, 118 low-risk (LR) and 372 high-risk (HR) drivers qualified for and consented to participate in a 2-month treatment period followed by 12 monthly recordings of driving mishaps. To address aim B, HR participants were randomized to receive either routine care (RC) or the online intervention "DiabetesDriving.com" (DD.com). Half of the DD.com participants received a motivational interview (MI) at the beginning and end of the treatment period to boost participation and efficacy. All of the LR participants were assigned to RC. In both studies, the primary outcome variable was driving mishaps.ResultsRelated to aim A, in study 1, the RADD demonstrated 61% sensitivity and 75% specificity. Participants in the upper third of the RADD distribution (HR), compared with those in the lower third (LR), reported 3.03 vs. 0.87 mishaps/driver/year, respectively (P < 0.001). In study 2, HR and LR participants receiving RC reported 4.3 and 1.6 mishaps/driver/year, respectively (P < 0.001). Related to aim B, in study 2, MIs did not enhance participation or efficacy, so the DD.com and DD.com + MI groups were combined. DD.com participants reported fewer hypoglycemia-related driving mishaps than HR participants receiving RC (P = 0.01), but more than LR participants receiving RC, reducing the difference between the HR and LR participants receiving RC by 63%. HR drivers differed from LR drivers at baseline across a variety of hypoglycemia and driving parameters.ConclusionsThe RADD identified higher-risk drivers, and identification seemed relatively stable across time, samples, and procedures. This 11-item questionnaire could inform patients at higher risk, and their clinicians, that they should take preventive steps to reduce driving mishaps, which was accomplished in aim B using DD.com.

Project description:Since the end of the nineteenth century, socio-economic changes have greatly altered the Central European landscape and the structural and functional quality of habitats. Urban sprawl areas have appeared, a reduction of multiple forest uses has resulted in the densification of forests and agricultural land use has changed fundamentally through specialisation and intensification. Many of these changes affect biodiversity. To determine the important drivers of spatio-temporal dynamics of the population of 28 game species, we first considered a total of 130 potential explanatory variables. Second, we aggregated the main drivers of single-species models for habitat guilds. Third, we evaluated the results to aid in the development and implementation of mitigation measures for different ecoregions. We used harvest data as a surrogate for population density from 1875 to 2014 in South Tyrol, Italy. In generalised linear models, we used environmental characteristics such as climate, landscape diversity and structures, land cover, hunting, wildlife diseases, competition and predation, land-use type, and intensity (including pesticide use) as explanatory variables to predict the spatio-temporal dynamics of game species. The important drivers are land use and management changes (intensification in the agriculturally favourable areas, extensification or abandonment in the unfavourable areas) as well as associated changes in the landscape features, diversity and structure, and hunting management. Climatic variables, interspecific competition and diseases only play a subordinate role. The dynamics of the habitat guilds and their drivers provide concrete indications for measures to maintain or improve the habitat quality for the investigated species. Particularly important are transfer payments to ensure extensive agricultural use, increasingly through the takeover of personnel costs, but also for the installation of an independent body that monitors and evaluates the effectiveness of the measures.

Dataset Information

Predicting Risky and Aggressive Driving Behavior among Taxi Drivers: Do Spatio-Temporal Attributes Matter?

Publications

Predicting Risky and Aggressive Driving Behavior among Taxi Drivers: Do Spatio-Temporal Attributes Matter?

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets