Individual driver risk assessment using naturalistic driving data

Guo, Feng; Fang, Youjia

doi:10.1016/j.aap.2012.06.014

Cited by 215 publications

(103 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Previous studies have explored these factors that influence the risk of driving behavior, weather, road environment, and driver demographics to provide Therefore, researchers have started paying attention to the potentiality of using near-crash data to study the significant factors related to driving risk to propose suggestions and recommendations to transportation regulators and enhance traffic safety. Guo and Fang [16] presented a method for assessing the driving risk of individual drivers using naturalistic driving data. A negative binomial regression model was used to examine the significant risk factors.…”

Section: Introductionmentioning

confidence: 99%

“…This study evaluated driving risk by the contributing factors of near-crash events (similar to Wang et al [17] and Guo et al [16]) rather than analyzing driving risk by crash data (similar to Chang et al [18]). In crash data, if severity levels are precise and identified in advance, then one needs only to select the proper regression model and load the severity of crashes as the dependent variable and other factors as the independent variables.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Evaluating the Driving Risk of Near-Crash Events Using a Mixed-Ordered Logit Model

et al. 2018

View full text Add to dashboard Cite

Abstract:With the considerable increase in ownership of motor vehicles, traffic crashes have become a challenge. This paper presents a study of naturalistic driving conducted to collect driving data. The experiments were performed on different road types in the city of Wuhan in China. The collected driving data were used to develop a near-crash database, which covers driving behavior, near-crash factors, driving environment, time, demographics, and experience. A new definition of near-crash events is also proposed. The new definition considers potential risks in driving behavior, such as braking pressure, time headway, and deceleration. A clustering analysis was carried out through a K-means algorithm to classify near-crash events based on their risk level. In addition, a mixed-ordered logit model was used to examine the contributing factors associated with the driving risk of near-crash events. The results indicate that ten factors significantly affect the driving risk of near-crash events: deceleration average, vehicle kinetic energy, near-crash causes, congestion on roads, time of day, driving miles, road types, weekend, age, and experience. The findings may be used by transportation planners to understand the factors that influence driving risk and may provide valuable insights and helpful suggestions for improving transportation rules and reducing traffic collisions thus making roads safer.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Evaluating the Driving Risk of Near-Crash Events Using a Mixed-Ordered Logit Model

et al. 2018

View full text Add to dashboard Cite

show abstract

“…For example, with the wide application of car GPS, individual vehicle could be easily turned into a data point with many dimensions (e.g., various vehicle features). Instead of running intentionally designed driving scenarios, investigators could investigate important traffic issues under a naturalistic driving context [68,69]. Visual data analytical techniques could be applied to explore and visualize the interactive effects of safety-threatening factors, such as hand-held cell phone [70], sleep habits [71], hormone responses [72], age [73], and other distraction factors.…”

Section: Chapter 5 Conclusion and Future Workmentioning

confidence: 99%

Be the Data: Embodied Visual Analytics

Chen¹,

Self

House

et al. 2018

IEEE Trans. Learning Technol.

View full text Add to dashboard Cite

(ABSTRACT)With the rise of big data, it is becoming increasingly important to educate students about data analytics. In particular, students without a strong mathematical background usually have an unenthusiastic attitude towards high-dimensional data and find it challenging to understand relevant complex analytical methods, such as dimension reduction. In this thesis, we present an embodied approach for visual analytics designed to teach students exploring alternative 2D projections of high dimensional data points using weighted multidimensional scaling. We proposed a novel concept, Be the Data, and its application to explore the possibilities of using human's embodied resources to learn from high dimensional data. In our system, each student embodies a data point and the position of students in a physical space represents a 2D projection of the high-dimensional data.Students physically moves in a room with respect to others to interact with alternative projections and receive visual feedback. We conducted educational workshops with students inexperienced in relevant data analytical methods. Our findings indicate that the students were able to learn about high-dimensional data and data analysis process despite their low level of knowledge about the complex analytical methods. Similarly, we applied Be the Data to social meetings. We used the same techniques to analyze and display social-cluster related information to facilitate social interactions in real time. Participants information 323.2 A quantitative summary of students' understanding about the key concepts (i.e., variable, relative distance, dimension reduction, data exploration), interests, and confidence towards learning high-dimensional data before and after the workshop. DR stands for dimension reduction. Column 3 and 4 are observed proportion of correct answers for the pre and post surveys. Column 5 and 6 are the expected difference and the credible interval for the difference in proportions. Column 7 are the p-values from a two-tailed two sample t-test. The * in column 6 and 7 flags questions when there is important difference in pre and post. 373.3 A summary of students' embodied interaction during the activity 48 4.1 A portion of the high-dimensional dataset that describes participants in one social meeting.

show abstract

“…For this aim, different spatial and non-spatial analysis methods are used for determining common properties of the spatial events. In this context, [20], [21], [22] and [23] used different clustering methods for determining common properties of different spatial events. Mapping of spatial events according with common properties is also significant for estimating types and effects of future events.…”

Section: Introductionmentioning

confidence: 99%

Using K-Means and K-Medoids Methods for Multivariate Mapping

Selvi

Çağlar²

2016

International Journal of Applied Mathematics, Electronics and Computers

View full text Add to dashboard Cite

Multivariate mapping is the visual exploration of multiple attributes using a map or data reduction technique. The simultaneous display of sometimes multiple features and their respective multivariate attributes allows for estimation of the degree or spatial pattern of cross-correlation between attributes. Multivariate mapping integrates computational, visual, and cartographic methods to develop a visual approach for exploring and understanding spatiotemporal and multivariate patterns. More than one attribute can be visually explored and symbolized using numerous statistical classification systems or data reduction techniques. In this sense, clustering analysis methods can be used for multivariate mapping. k-means and k-medoids methods which are non-hierarchical clustering analysis methods were analyzed in this study. The aim of this study is to determine the success of the spatial analysis of the multivariate maps produced by these methods. For this aim, classes and multivariate maps created with these methods from traffic accident data of two different years in Turkey were presented. In addition usability of such maps in risk management and planning was discussed.

show abstract

Individual driver risk assessment using naturalistic driving data

Cited by 215 publications

References 18 publications

Evaluating the Driving Risk of Near-Crash Events Using a Mixed-Ordered Logit Model

Evaluating the Driving Risk of Near-Crash Events Using a Mixed-Ordered Logit Model

Be the Data: Embodied Visual Analytics

Using K-Means and K-Medoids Methods for Multivariate Mapping

Contact Info

Product

Resources

About