TRB Pubsindex
Text Size:

Title:

Topic Models from Crash Narrative Reports of Motorcycle Crash Causation Study

Accession Number:

01764163

Record Type:

Component

Availability:

Find a library where document is available


Order URL: http://worldcat.org/issn/03611981

Abstract:

The Motorcycle Crash Causation Study (MCCS) is a matched case-control study that contains a very wide list of crash contributing factors associated with motorcycle crash occurrences. It contains information such as motorcycle information, rider information, and associated trip information. This study also provides crash narrative information that presents an in-depth narrative discussion of the crash causation. Because of the plethora of information, it is critical to investigate MCCS-related data. Some studies examined the structured information in MCCS datasets. There is no in-depth study that has examined the unstructured textual contents in the MCCS data. This study aims to mitigate this research gap by applying different natural language processing tools (e.g., text mining, topic modeling). Fatal and non-fatal crash narratives are clustered separately to gain insights pertaining to the injury level. The findings of this study will contribute to the ongoing studies on MCCS to better understand the crash causation mechanism associated with motorcycle crashes.

Supplemental Notes:

© National Academy of Sciences: Transportation Research Board 2021.

Report/Paper Numbers:

TRBAM-21-03150

Language:

English

Authors:

Das, Subasish
Dutta, Anandi
Tsapakis, Ioannis

Pagination:

pp 449-462

Publication Date:

2021-9

Serial:

Transportation Research Record: Journal of the Transportation Research Board

Volume: 2675
Issue Number: 9
Publisher: Sage Publications, Incorporated
ISSN: 0361-1981
EISSN: 2169-4052
Serial URL: http://journals.sagepub.com/home/trr

Media Type:

Digital/other

Features:

Figures; References (49) ; Tables

Subject Areas:

Highways; Safety and Human Factors; Vehicles and Equipment

Files:

TRIS, TRB, ATRI

Created Date:

Dec 23 2020 11:21AM