Using Twitter data for transit performance assessment: a framework for evaluating transit riders’ opinions about quality of service

N. Nima Haghighi, Xiaoyue Cathy Liu, Ran Wei, WenWen Li, Hu Shao

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Social media platforms such as Facebook, Instagram, and Twitter have drastically altered the way information is generated and disseminated. These platforms allow their users to report events and express their opinions toward these events. The profusion of data generated through social media has proved to have the potential for improving the efficiency of existing traffic management systems and transportation analytics. This study complements existing literature by proposing a framework to evaluate transit riders’ opinion about quality of transit service using Twitter data. Although previous studies used keyword search to extract transit-related tweets, the extracted tweets can still be noisy and might not be relevant to transit quality of service at all. In this study, we leverage topic modeling, an unsupervised machine learning technique, to sift tweets that are relevant to the actual user experience of the transit system. Sentiment analysis is further performed based on the tweet-per-topic index we developed, to gauge transit riders’ feedback and explore the underlying reasons causing their dissatisfaction on the service. This framework can be potentially quite useful to transit agencies for user-oriented analysis and to assist with investment decision making.

Original languageEnglish (US)
Pages (from-to)363-377
Number of pages15
JournalPublic Transport
Volume10
Issue number2
DOIs
StatePublished - Aug 1 2018

Fingerprint

twitter
performance assessment
Gages
Learning systems
Quality of service
Decision making
social media
Feedback
traffic management system
event
facebook
decision making
efficiency
learning
Twitter
Performance assessment
experience

Keywords

  • Latent Dirichlet allocation (LDA)
  • Quality of transit service
  • Sentiment analysis
  • Topic modeling
  • Transit service performance

ASJC Scopus subject areas

  • Information Systems
  • Transportation
  • Mechanical Engineering
  • Management Science and Operations Research

Cite this

Using Twitter data for transit performance assessment : a framework for evaluating transit riders’ opinions about quality of service. / Haghighi, N. Nima; Liu, Xiaoyue Cathy; Wei, Ran; Li, WenWen; Shao, Hu.

In: Public Transport, Vol. 10, No. 2, 01.08.2018, p. 363-377.

Research output: Contribution to journalArticle

@article{f11672f5cbbc400983350f0ecddabd9f,
title = "Using Twitter data for transit performance assessment: a framework for evaluating transit riders’ opinions about quality of service",
abstract = "Social media platforms such as Facebook, Instagram, and Twitter have drastically altered the way information is generated and disseminated. These platforms allow their users to report events and express their opinions toward these events. The profusion of data generated through social media has proved to have the potential for improving the efficiency of existing traffic management systems and transportation analytics. This study complements existing literature by proposing a framework to evaluate transit riders’ opinion about quality of transit service using Twitter data. Although previous studies used keyword search to extract transit-related tweets, the extracted tweets can still be noisy and might not be relevant to transit quality of service at all. In this study, we leverage topic modeling, an unsupervised machine learning technique, to sift tweets that are relevant to the actual user experience of the transit system. Sentiment analysis is further performed based on the tweet-per-topic index we developed, to gauge transit riders’ feedback and explore the underlying reasons causing their dissatisfaction on the service. This framework can be potentially quite useful to transit agencies for user-oriented analysis and to assist with investment decision making.",
keywords = "Latent Dirichlet allocation (LDA), Quality of transit service, Sentiment analysis, Topic modeling, Transit service performance",
author = "Haghighi, {N. Nima} and Liu, {Xiaoyue Cathy} and Ran Wei and WenWen Li and Hu Shao",
year = "2018",
month = "8",
day = "1",
doi = "10.1007/s12469-018-0184-4",
language = "English (US)",
volume = "10",
pages = "363--377",
journal = "Public Transport",
issn = "1613-7159",
publisher = "Springer Verlag",
number = "2",

}

TY - JOUR

T1 - Using Twitter data for transit performance assessment

T2 - a framework for evaluating transit riders’ opinions about quality of service

AU - Haghighi, N. Nima

AU - Liu, Xiaoyue Cathy

AU - Wei, Ran

AU - Li, WenWen

AU - Shao, Hu

PY - 2018/8/1

Y1 - 2018/8/1

N2 - Social media platforms such as Facebook, Instagram, and Twitter have drastically altered the way information is generated and disseminated. These platforms allow their users to report events and express their opinions toward these events. The profusion of data generated through social media has proved to have the potential for improving the efficiency of existing traffic management systems and transportation analytics. This study complements existing literature by proposing a framework to evaluate transit riders’ opinion about quality of transit service using Twitter data. Although previous studies used keyword search to extract transit-related tweets, the extracted tweets can still be noisy and might not be relevant to transit quality of service at all. In this study, we leverage topic modeling, an unsupervised machine learning technique, to sift tweets that are relevant to the actual user experience of the transit system. Sentiment analysis is further performed based on the tweet-per-topic index we developed, to gauge transit riders’ feedback and explore the underlying reasons causing their dissatisfaction on the service. This framework can be potentially quite useful to transit agencies for user-oriented analysis and to assist with investment decision making.

AB - Social media platforms such as Facebook, Instagram, and Twitter have drastically altered the way information is generated and disseminated. These platforms allow their users to report events and express their opinions toward these events. The profusion of data generated through social media has proved to have the potential for improving the efficiency of existing traffic management systems and transportation analytics. This study complements existing literature by proposing a framework to evaluate transit riders’ opinion about quality of transit service using Twitter data. Although previous studies used keyword search to extract transit-related tweets, the extracted tweets can still be noisy and might not be relevant to transit quality of service at all. In this study, we leverage topic modeling, an unsupervised machine learning technique, to sift tweets that are relevant to the actual user experience of the transit system. Sentiment analysis is further performed based on the tweet-per-topic index we developed, to gauge transit riders’ feedback and explore the underlying reasons causing their dissatisfaction on the service. This framework can be potentially quite useful to transit agencies for user-oriented analysis and to assist with investment decision making.

KW - Latent Dirichlet allocation (LDA)

KW - Quality of transit service

KW - Sentiment analysis

KW - Topic modeling

KW - Transit service performance

UR - http://www.scopus.com/inward/record.url?scp=85052294717&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85052294717&partnerID=8YFLogxK

U2 - 10.1007/s12469-018-0184-4

DO - 10.1007/s12469-018-0184-4

M3 - Article

AN - SCOPUS:85052294717

VL - 10

SP - 363

EP - 377

JO - Public Transport

JF - Public Transport

SN - 1613-7159

IS - 2

ER -