Young driver fatal motorcycle accident analysis by jointly maximizing accuracy and information. (August 2019)

• Record Type: Journal Article
• Authors: Halbersberg, Dan; Lerner, Boaz
• Is Part Of: Accident analysis and prevention. Volume 129(2019)
• Page Start: 350
• Page End: 361
• Publication Date: 2019-08
• Subjects: Young drivers -- Fatal accidents -- Motorcycle -- Machine learning -- Information measure -- Prediction -- Bayesian network -- Key factors; Accidents -- Prevention -- Periodicals; Accident Prevention -- Periodicals; Accidents -- Prévention -- Périodiques; 363.106
• DOI: 10.1016/j.aap.2019.04.016
• Languages: English
• ISSNs: 0001-4575

Abstract

Highlights: Predicting fatal motorcycle accidents (FMAs) of young drivers (YDs) and identifying key factors. FMA prediction is not accurate as classifiers focus on minor rather than fatal accidents. Classifiers provide no information about error distribution and insensitive to error severity. A proposed measure maximizes accuracy + information; sensitive to error distribution & severity. Using the measure, we predicted FMAs better and identified fatal accident key factors. Abstract: While young drivers (YDs) constitute ∼10% of the driver population, their fatality rate in motorcycle accidents is up to three times higher. Thus, we are interested in predicting fatal motorcycle accidents (FMAs), and in identifying their key factors and possible causes. Accurate prediction of YD FMAs from data by risk minimization using the 0/1 loss function (i.e., the ordinary classification accuracy) cannot be guaranteed because these accidents are only ∼1% of all YD motorcycle accidents, and classifiers tend to focus on the majority class of minor accidents at the expense of the minority class of fatal ones. Also, classifiers are usually uninformative (providing no information about the distribution of misclassifications), insensitive to error severity (making no distinction between misclassification of fatal accidents as severe or minor), and limited in identifying key factors. We propose to use an information measure (IM) that jointly maximizes accuracy and information and is sensitive to the error distribution and severity. Using a database of ∼3600 motorcycle accidents, a Bayesian network classifier optimized by IM predicted FMAs better than classifiers maximizing accuracy or other predictive or information measures, and identified fatal accident key factors and causal relations.