DOI

10.17077/drivingassessment.1696

Location

Santa Fe, New Mexico, USA

Date

26-6-2019

Session

Session 5 – Methods and Data Analysis

Abstract

A wealth of literature has shown the predictive and preventive utility of the Big Five personality traits model (BIG5) for various kinds of unsafe driving. However, the commonly used method for BIG5 measurement requires subjects to answer long and stressful questionnaires, making its applicability limited. In this paper, we study the potential for predicting a driver's BIG5 traits from his/her daily driving behavior. We collected naturalistic driving data on (A) car usage behavior (driving frequency, distance, duration, etc.) and (B) driving operation behavior (operation of steering wheel, accelerator and brake pedal, etc.) from 140 Japanese subjects over two months. By analyzing the data while focusing on various specific driving conditions, we were able to find features which significantly correlate with BIG5 traits from both (A) and (B). In the evaluation, the features we found predicted whether the traits scores are above μ + σ or below μ - σ (μ: average, σ: standard deviation) at an accuracy of ROC-AUC 0.62~0.85, confirming the potential for predicting BIG5 traits from daily driving behavior.

Rights

Copyright © 2019 the author(s)

DC Citation

Proceedings of the Tenth International Driving Symposium on Human Factors in Driver Assessment, Training and Vehicle Design, 24-27 June 2019, Santa Fe, New Mexico. Iowa City, IA: Public Policy Center, of Iowa, 2019: 203-209.

Share

COinS
 
Jun 26th, 12:00 AM

Predicting a Driver's Personality from Daily Driving Behavior

Santa Fe, New Mexico, USA

A wealth of literature has shown the predictive and preventive utility of the Big Five personality traits model (BIG5) for various kinds of unsafe driving. However, the commonly used method for BIG5 measurement requires subjects to answer long and stressful questionnaires, making its applicability limited. In this paper, we study the potential for predicting a driver's BIG5 traits from his/her daily driving behavior. We collected naturalistic driving data on (A) car usage behavior (driving frequency, distance, duration, etc.) and (B) driving operation behavior (operation of steering wheel, accelerator and brake pedal, etc.) from 140 Japanese subjects over two months. By analyzing the data while focusing on various specific driving conditions, we were able to find features which significantly correlate with BIG5 traits from both (A) and (B). In the evaluation, the features we found predicted whether the traits scores are above μ + σ or below μ - σ (μ: average, σ: standard deviation) at an accuracy of ROC-AUC 0.62~0.85, confirming the potential for predicting BIG5 traits from daily driving behavior.