Variations on a theme

Topic modeling of naturalistic driving data

Elease McLaurin, Anthony D. McDonald, John D. Lee, Nazan Aksan, Jeffrey Dawson, Jon Tippin, Matthew Rizzo

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

This paper introduces Probabilistic Topic Modeling (PTM) as a promising approach to naturalistic driving data analyses. Naturalistic driving data present an unprecedented opportunity to understand driver behavior. Novel strategies are needed to achieve a more complete picture of these datasets than is provided by the local event-based analytic strategy that currently dominates the field. PTM is a text analysis method for uncovering word-based themes across documents. In this application, documents were represented by drives and words were created from speed and acceleration data using Symbolic Aggregate approximation (SAX). A twenty-topic Latent Dirichlet Allocation (LDA) topic model was developed using words from 10,705 documents (real-world drives) by 26 drivers. The resulting LDA model clustered the drives into meaningful topics. Topic membership probabilities were successfully used as features in subsequent analyses to differentiate between healthy drivers and those suffering from Obstructive Sleep Apnea.

Original languageEnglish (US)
Title of host publicationProceedings of the Human Factors and Ergonomics Society
PublisherHuman Factors an Ergonomics Society Inc.
Pages2107-2111
Number of pages5
Volume2014-January
ISBN (Print)9780945289456
DOIs
StatePublished - 2014
Externally publishedYes
Event58th International Annual Meeting of the Human Factors and Ergonomics Society, HFES 2014 - Chicago, United States
Duration: Oct 27 2014Oct 31 2014

Other

Other58th International Annual Meeting of the Human Factors and Ergonomics Society, HFES 2014
CountryUnited States
CityChicago
Period10/27/1410/31/14

Fingerprint

driver
text analysis
sleep
event
Sleep

ASJC Scopus subject areas

  • Human Factors and Ergonomics

Cite this

McLaurin, E., McDonald, A. D., Lee, J. D., Aksan, N., Dawson, J., Tippin, J., & Rizzo, M. (2014). Variations on a theme: Topic modeling of naturalistic driving data. In Proceedings of the Human Factors and Ergonomics Society (Vol. 2014-January, pp. 2107-2111). Human Factors an Ergonomics Society Inc.. https://doi.org/10.1177/1541931214581443

Variations on a theme : Topic modeling of naturalistic driving data. / McLaurin, Elease; McDonald, Anthony D.; Lee, John D.; Aksan, Nazan; Dawson, Jeffrey; Tippin, Jon; Rizzo, Matthew.

Proceedings of the Human Factors and Ergonomics Society. Vol. 2014-January Human Factors an Ergonomics Society Inc., 2014. p. 2107-2111.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

McLaurin, E, McDonald, AD, Lee, JD, Aksan, N, Dawson, J, Tippin, J & Rizzo, M 2014, Variations on a theme: Topic modeling of naturalistic driving data. in Proceedings of the Human Factors and Ergonomics Society. vol. 2014-January, Human Factors an Ergonomics Society Inc., pp. 2107-2111, 58th International Annual Meeting of the Human Factors and Ergonomics Society, HFES 2014, Chicago, United States, 10/27/14. https://doi.org/10.1177/1541931214581443
McLaurin E, McDonald AD, Lee JD, Aksan N, Dawson J, Tippin J et al. Variations on a theme: Topic modeling of naturalistic driving data. In Proceedings of the Human Factors and Ergonomics Society. Vol. 2014-January. Human Factors an Ergonomics Society Inc. 2014. p. 2107-2111 https://doi.org/10.1177/1541931214581443
McLaurin, Elease ; McDonald, Anthony D. ; Lee, John D. ; Aksan, Nazan ; Dawson, Jeffrey ; Tippin, Jon ; Rizzo, Matthew. / Variations on a theme : Topic modeling of naturalistic driving data. Proceedings of the Human Factors and Ergonomics Society. Vol. 2014-January Human Factors an Ergonomics Society Inc., 2014. pp. 2107-2111
@inproceedings{d25612a12fa840a8adb73a2f2312fb74,
title = "Variations on a theme: Topic modeling of naturalistic driving data",
abstract = "This paper introduces Probabilistic Topic Modeling (PTM) as a promising approach to naturalistic driving data analyses. Naturalistic driving data present an unprecedented opportunity to understand driver behavior. Novel strategies are needed to achieve a more complete picture of these datasets than is provided by the local event-based analytic strategy that currently dominates the field. PTM is a text analysis method for uncovering word-based themes across documents. In this application, documents were represented by drives and words were created from speed and acceleration data using Symbolic Aggregate approximation (SAX). A twenty-topic Latent Dirichlet Allocation (LDA) topic model was developed using words from 10,705 documents (real-world drives) by 26 drivers. The resulting LDA model clustered the drives into meaningful topics. Topic membership probabilities were successfully used as features in subsequent analyses to differentiate between healthy drivers and those suffering from Obstructive Sleep Apnea.",
author = "Elease McLaurin and McDonald, {Anthony D.} and Lee, {John D.} and Nazan Aksan and Jeffrey Dawson and Jon Tippin and Matthew Rizzo",
year = "2014",
doi = "10.1177/1541931214581443",
language = "English (US)",
isbn = "9780945289456",
volume = "2014-January",
pages = "2107--2111",
booktitle = "Proceedings of the Human Factors and Ergonomics Society",
publisher = "Human Factors an Ergonomics Society Inc.",

}

TY - GEN

T1 - Variations on a theme

T2 - Topic modeling of naturalistic driving data

AU - McLaurin, Elease

AU - McDonald, Anthony D.

AU - Lee, John D.

AU - Aksan, Nazan

AU - Dawson, Jeffrey

AU - Tippin, Jon

AU - Rizzo, Matthew

PY - 2014

Y1 - 2014

N2 - This paper introduces Probabilistic Topic Modeling (PTM) as a promising approach to naturalistic driving data analyses. Naturalistic driving data present an unprecedented opportunity to understand driver behavior. Novel strategies are needed to achieve a more complete picture of these datasets than is provided by the local event-based analytic strategy that currently dominates the field. PTM is a text analysis method for uncovering word-based themes across documents. In this application, documents were represented by drives and words were created from speed and acceleration data using Symbolic Aggregate approximation (SAX). A twenty-topic Latent Dirichlet Allocation (LDA) topic model was developed using words from 10,705 documents (real-world drives) by 26 drivers. The resulting LDA model clustered the drives into meaningful topics. Topic membership probabilities were successfully used as features in subsequent analyses to differentiate between healthy drivers and those suffering from Obstructive Sleep Apnea.

AB - This paper introduces Probabilistic Topic Modeling (PTM) as a promising approach to naturalistic driving data analyses. Naturalistic driving data present an unprecedented opportunity to understand driver behavior. Novel strategies are needed to achieve a more complete picture of these datasets than is provided by the local event-based analytic strategy that currently dominates the field. PTM is a text analysis method for uncovering word-based themes across documents. In this application, documents were represented by drives and words were created from speed and acceleration data using Symbolic Aggregate approximation (SAX). A twenty-topic Latent Dirichlet Allocation (LDA) topic model was developed using words from 10,705 documents (real-world drives) by 26 drivers. The resulting LDA model clustered the drives into meaningful topics. Topic membership probabilities were successfully used as features in subsequent analyses to differentiate between healthy drivers and those suffering from Obstructive Sleep Apnea.

UR - http://www.scopus.com/inward/record.url?scp=84957671415&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84957671415&partnerID=8YFLogxK

U2 - 10.1177/1541931214581443

DO - 10.1177/1541931214581443

M3 - Conference contribution

SN - 9780945289456

VL - 2014-January

SP - 2107

EP - 2111

BT - Proceedings of the Human Factors and Ergonomics Society

PB - Human Factors an Ergonomics Society Inc.

ER -