Vehicle Crash Analysis

Project Overview

The Virginia Department of Transportation (VDOT) has archived a large number of FR300 reports documenting vehicular accidents in the Commonwealth, and thousands of additional reports accumulate each month. Safety analysts use this valuable resource to understand the patterns and factors of vehicular accidents. This project aims to refine and expand these capabilities using semantic analysis methods to extract information from the free-text narrative descriptions provided in most FR300s. We are focusing on two areas of interest to the Traffic Engineering Division regarding the FR300 reports: quality control and anomaly detection. We are designing, implementing, and validating proof-of-concept semantic analysis algorithms that support these areas of interest. This work will set the stage for future, large-scale projects that enhance the utility of the FR300 database through automatic, scalable analysis of crash report narratives. [1]

Members


References

  1. [gerber2013automatic] Gerber, M., and L. Tang, "Automatic Quality Control of Transportation Reports using Statistical Language Processing", IEEE Transactions on Intelligent Transportation Systems, vol. 14, issue 4, pp. 1681-1689, 12/2013.