Probabilistic Determination of Crash Locations in a Road Network with Imperfect Data

Abstract

The limited quality of location data poses a major problem to those who want to use such data for research or safety management. The proportion of crashes unassigned to specific road locations is considerable. This fact is sometimes overlooked because crash database queries return crashes assignable to locations but do not issue warnings about crashes that are not assignable. Without reliable location information, crash data are of limited use for more refined safety research and even for basic road safety management. Probabilistic linking techniques are frequently chosen to link medical, census, and other population records, mainly for medical research but also for security concerns and market studies. This paper presents a first application of the probabilistic method of assigning crashes to roads through linking road crash records with road inventory records. Because of missing or incorrect data, multiple locations are candidates, and probabilistic linking offers a solution by selecting the highest-likelihood locations. This paper uses the Bayesian approach to link data in a version specialized for linking one-to-many records.

Get full access to this article

View all access options for this article.

References

Fellegi

I. P.

, and Sunter

A. P.

A Theory for Record Linkage.

Journal of the American Statistical Association, Vol. 64, 1969, pp. 1183–1210.

Dey

, Sarkar

, and De

A Probabilistic Decision Model for Entity Matching in Heterogeneous Databases.

Management Science, Vol. 44, No. 10, 1998, pp. 1379–1395.

Dey

, Sarkar

, and De

A Distance-Based Approach to Entity Reconciliation in Heterogeneous Databases.

IEEE Transactions on Knowledge and Data Engineering, Vol. 14, No. 3, 2002, pp. 567–582.

Dempster

A. P.

, Laird

N. M.

, and Rubin

D. B.

Maximum Likelihood from Incomplete Data via the EM Algorithm.

Journal of the Royal Statistical Society, Vol. 39, 1977, pp. 1–38.

Jaro

M. A.

Advances in Record-Linkage Methodology as Applied to Matching the 1985 Census of Tampa, Florida.

Journal of the American Statistical Association, Vol. 89, 1989, pp. 414–420.

Winkler

W. E.

Frequency-Based Matching in the Fellegi–Sunter Model of Record Linkage.

Proceedings of Survey Research Methods Section, American Statistical Association, 1989, pp. 778–783.

Winkler

W. E.

String Comparator Metrics and Enhanced Decision Rules in the Fellegi–Sunter Model of Record Linkage.

Proceedings of Survey Research Methods Section, American Statistical Association, 1990, pp. 472–477.

Winkler

W. E.

, and Thibeaudeau

An Application of the Fellegi–Sunter Model of Record Linkage to the 1990 U.S. Decennial Census. Statistical Research Division Report, U.S. Bureau of the Census, 1991.

Grannis

S. J.

, Overhage

J. M.

, Hui

, and McDonald

C. J.

Analysis of a Probabilistic Record Linkage Technique Without Human Review.

Proceedings of the American Medical Informatics Association Annual Symposium, 2003, pp. 259–263.

10.

Yancey

W. E.

Frequency-Dependent Probability Measures for Record Linkage.

Proceedings of Survey Research Methods Section, American Statistical Association, 2000, pp. 752–757.

11.

, Baxter

, Vickers

, and Rainsford

Record Linkage: Current Practice and Future Directions. Technical Report 03/83. Commonwealth Scientific and Industrial Research Organisation, Mathematical and Information Sciences, Canberra, Australia, 2003.

12.

Elfeky

M. G.

, Verykios

V. S.

, and Elmagarmid

A. K.

TAILOR: A Record Linkage Toolbox. Proc., 18th International Conference on Data Engineering, IEEE, 2002.

13.

Johnson

Technical Issues Related to the Probabilistic Linkage of Population-Based Crash and Injury Data.

Record Linkage Techniques— 1997: Proceedings of an International Workshop and Exposition, National Research Council, Washington, D.C., 1999, pp. 222–226.

14.

McGlincy

M. H.

A Bayesian Record Linkage Methodology for Multiple Imputation of Missing Links. Survey Research Methods Section, American Statistical Association, 2004. www.amstat.org/sections/SRMS/Proceedings/y2004/files/Jsm2004-000683.pdf.

15.

Larsen

M. D.

Modeling Issues and the Use of Experience in Record Linkage.

Record Linkage Techniques—1997: Proceedings of an International Workshop and Exposition, National Research Council, Washington, D.C., 1999, pp. 95–105.

16.

Larsen

M. D.

Hierarchical Bayesian Record Linkage and Regression in Linked Files.

Proceedings of American Mathematical Society–Institute of Mathematical Statistics–Society for Industrial and Applied Mathematics Joint Summer Research Conference on Machine Learning, Statistics, and Discovery, Snowbird, Utah, 2003.

17.

Lahiri

, and Larsen

M. D.

Regression Analysis with Linked Data.

Journal of the American Statistical Association, Vol. 100, 2005, pp. 222–230.

18.

Tarko

A. P.

, Iqbal

M. A.

, Inerowicz

, Liang

, Panicker

, and Ramos

Safety Conscious Planning in Indiana: Predicting Safety Benefits in Corridor Studies. Joint Transportation Research Program Draft Final Report, FHWA/IN/JTRP-2006/21. School of Civil Engineering, Purdue University, West Lafayette, Ind., 2007.