A Comparison of LSA and LDA for the Analysis of Railroad Accident Text

Trefor Williams; John Betak

doi:10.1016/j.procs.2018.04.017

A Comparison of LSA and LDA for the Analysis of Railroad Accident Text

Trefor Williams, John Betak

School of Engineering, Civil & Environmental Engineering

Research output: Contribution to journal › Conference article › peer-review

41 Scopus citations

Abstract

Latent Semantic Analysis (LSA) and Latent Dirichlet Allocation(LDA) were used to identify themes in a database of text about railroad equipment accidents maintained by the Federal Railroad Administration in the United States. These text mining techniques use different mechanisms to identify topics. LDA and LSA identified switching accidents, hump yard accidents and grade crossing accidents as major accident type topics. LSA identified accidents with track maintenance equipment as a topic. Both text mining models identified accidents with tractor-trailer highway trucks as a particular problem at grade crossings. It was found that the use of the two techniques was complementary, with more accident topics identified than with the use of a single method.

Original language	English (US)
Pages (from-to)	98-102
Number of pages	5
Journal	Procedia Computer Science
Volume	130
DOIs	https://doi.org/10.1016/j.procs.2018.04.017
State	Published - 2018
Event	9th International Conference on Ambient Systems, Networks and Technologies, ANT 2018 - Porto, Indonesia Duration: May 8 2018 → May 11 2018

All Science Journal Classification (ASJC) codes

General Computer Science

Keywords

accidents
railroad
text mining

Access to Document

10.1016/j.procs.2018.04.017

Cite this

@article{9a5a8ebbf4314da785227534aa477720,

title = "A Comparison of LSA and LDA for the Analysis of Railroad Accident Text",

abstract = "Latent Semantic Analysis (LSA) and Latent Dirichlet Allocation(LDA) were used to identify themes in a database of text about railroad equipment accidents maintained by the Federal Railroad Administration in the United States. These text mining techniques use different mechanisms to identify topics. LDA and LSA identified switching accidents, hump yard accidents and grade crossing accidents as major accident type topics. LSA identified accidents with track maintenance equipment as a topic. Both text mining models identified accidents with tractor-trailer highway trucks as a particular problem at grade crossings. It was found that the use of the two techniques was complementary, with more accident topics identified than with the use of a single method.",

keywords = "accidents, railroad, text mining",

author = "Trefor Williams and John Betak",

note = "Publisher Copyright: {\textcopyright} 2018 The Authors. Published by Elsevier B.V.; 9th International Conference on Ambient Systems, Networks and Technologies, ANT 2018 ; Conference date: 08-05-2018 Through 11-05-2018",

year = "2018",

doi = "10.1016/j.procs.2018.04.017",

language = "English (US)",

volume = "130",

pages = "98--102",

journal = "Procedia Computer Science",

issn = "1877-0509",

publisher = "Elsevier B.V.",

}

TY - JOUR

T1 - A Comparison of LSA and LDA for the Analysis of Railroad Accident Text

AU - Williams, Trefor

AU - Betak, John

PY - 2018

Y1 - 2018

N2 - Latent Semantic Analysis (LSA) and Latent Dirichlet Allocation(LDA) were used to identify themes in a database of text about railroad equipment accidents maintained by the Federal Railroad Administration in the United States. These text mining techniques use different mechanisms to identify topics. LDA and LSA identified switching accidents, hump yard accidents and grade crossing accidents as major accident type topics. LSA identified accidents with track maintenance equipment as a topic. Both text mining models identified accidents with tractor-trailer highway trucks as a particular problem at grade crossings. It was found that the use of the two techniques was complementary, with more accident topics identified than with the use of a single method.

AB - Latent Semantic Analysis (LSA) and Latent Dirichlet Allocation(LDA) were used to identify themes in a database of text about railroad equipment accidents maintained by the Federal Railroad Administration in the United States. These text mining techniques use different mechanisms to identify topics. LDA and LSA identified switching accidents, hump yard accidents and grade crossing accidents as major accident type topics. LSA identified accidents with track maintenance equipment as a topic. Both text mining models identified accidents with tractor-trailer highway trucks as a particular problem at grade crossings. It was found that the use of the two techniques was complementary, with more accident topics identified than with the use of a single method.

KW - accidents

KW - railroad

KW - text mining

UR - http://www.scopus.com/inward/record.url?scp=85051249391&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85051249391&partnerID=8YFLogxK

U2 - 10.1016/j.procs.2018.04.017

DO - 10.1016/j.procs.2018.04.017

M3 - Conference article

AN - SCOPUS:85051249391

SN - 1877-0509

VL - 130

SP - 98

EP - 102

JO - Procedia Computer Science

JF - Procedia Computer Science

T2 - 9th International Conference on Ambient Systems, Networks and Technologies, ANT 2018

Y2 - 8 May 2018 through 11 May 2018

ER -

A Comparison of LSA and LDA for the Analysis of Railroad Accident Text

Abstract

All Science Journal Classification (ASJC) codes

Keywords

Access to Document

Other files and links

Fingerprint

Cite this