Bayesian networks and traffic accident reconstruction

  • Authors:
  • Gary A. Davis;Jianping Pei

  • Affiliations:
  • University of Minnesota, Minneapolis, MN;University of Minnesota, Minneapolis, MN

  • Venue:
  • ICAIL '03 Proceedings of the 9th international conference on Artificial intelligence and law
  • Year:
  • 2003

Quantified Score

Hi-index 0.00

Visualization

Abstract

The attempt to draw rational conclusions about a road accident can be viewed as a problem in uncertain reasoning about a particular event, to which developments in the modeling of uncertain reasoning for artificial intelligence can be applied. Physical principles can be used to develop a structural model for the accident, and this model can then be combined with an expert assessment of prior uncertainty concerning the model's variables. Posterior probabilities, given evidence collected at the accident scene, can then be computed using Bayes theorem. Truth conditions for counterfactual claims about the accident can then be defined using a "possible worlds" semantics, and used to rigorously implement a "but for" test of whether or not a speed limit violation could be considered a cause of the accident.