Dynamic joint sentiment-topic model

Yulan He, Chenghua Lin, Wei Gao, Kam-Fai Wong

Research output: Contribution to journalArticle

39 Citations (Scopus)

Abstract

Social media data are produced continuously by a large and uncontrolled number of users. The dynamic nature of such data requires the sentiment and topic analysis model to be also dynamically updated, capturing the most recent language use of sentiments and topics in text. We propose a dynamic Joint Sentiment-Topic model (dJST) which allows the detection and tracking of views of current and recurrent interests and shifts in topic and sentiment. Both topic and sentiment dynamics are captured by assuming that the current sentiment-topic-specific word distributions are generated according to the word distributions at previous epochs. We study three different ways of accounting for such dependency information: (1) sliding window where the current sentiment-topic word distributions are dependent on the previous sentiment-topic-specific word distributions in the last S epochs; (2) skip model where history sentiment topic word distributions are considered by skipping some epochs in between; and (3) multiscale model where previous long- and short- timescale distributions are taken into consideration. We derive efficient online inference procedures to sequentially update the model with newly arrived data and show the effectiveness of our proposed model on the Mozilla add-on reviews crawled between 2007 and 2011.
Original languageEnglish
Article number6
Number of pages21
JournalACM Transactions on Intelligent Systems and Technology
Volume5
Issue number1
DOIs
Publication statusPublished - Dec 2013

Fingerprint

Model
Multiscale Model
Social Media
Sliding Window
Model Analysis
Time Scales
Update
Dependent
Language
Review
History
Text

Cite this

Dynamic joint sentiment-topic model. / He, Yulan; Lin, Chenghua; Gao, Wei; Wong, Kam-Fai.

In: ACM Transactions on Intelligent Systems and Technology, Vol. 5, No. 1, 6, 12.2013.

Research output: Contribution to journalArticle

He, Yulan ; Lin, Chenghua ; Gao, Wei ; Wong, Kam-Fai. / Dynamic joint sentiment-topic model. In: ACM Transactions on Intelligent Systems and Technology. 2013 ; Vol. 5, No. 1.
@article{08ec2b6b34224dd192e5e6c8e333efbb,
title = "Dynamic joint sentiment-topic model",
abstract = "Social media data are produced continuously by a large and uncontrolled number of users. The dynamic nature of such data requires the sentiment and topic analysis model to be also dynamically updated, capturing the most recent language use of sentiments and topics in text. We propose a dynamic Joint Sentiment-Topic model (dJST) which allows the detection and tracking of views of current and recurrent interests and shifts in topic and sentiment. Both topic and sentiment dynamics are captured by assuming that the current sentiment-topic-specific word distributions are generated according to the word distributions at previous epochs. We study three different ways of accounting for such dependency information: (1) sliding window where the current sentiment-topic word distributions are dependent on the previous sentiment-topic-specific word distributions in the last S epochs; (2) skip model where history sentiment topic word distributions are considered by skipping some epochs in between; and (3) multiscale model where previous long- and short- timescale distributions are taken into consideration. We derive efficient online inference procedures to sequentially update the model with newly arrived data and show the effectiveness of our proposed model on the Mozilla add-on reviews crawled between 2007 and 2011.",
author = "Yulan He and Chenghua Lin and Wei Gao and Kam-Fai Wong",
year = "2013",
month = "12",
doi = "10.1145/2542182.2542188",
language = "English",
volume = "5",
journal = "ACM Transactions on Intelligent Systems and Technology",
issn = "2157-6904",
publisher = "Association for Computing Machinery (ACM)",
number = "1",

}

TY - JOUR

T1 - Dynamic joint sentiment-topic model

AU - He, Yulan

AU - Lin, Chenghua

AU - Gao, Wei

AU - Wong, Kam-Fai

PY - 2013/12

Y1 - 2013/12

N2 - Social media data are produced continuously by a large and uncontrolled number of users. The dynamic nature of such data requires the sentiment and topic analysis model to be also dynamically updated, capturing the most recent language use of sentiments and topics in text. We propose a dynamic Joint Sentiment-Topic model (dJST) which allows the detection and tracking of views of current and recurrent interests and shifts in topic and sentiment. Both topic and sentiment dynamics are captured by assuming that the current sentiment-topic-specific word distributions are generated according to the word distributions at previous epochs. We study three different ways of accounting for such dependency information: (1) sliding window where the current sentiment-topic word distributions are dependent on the previous sentiment-topic-specific word distributions in the last S epochs; (2) skip model where history sentiment topic word distributions are considered by skipping some epochs in between; and (3) multiscale model where previous long- and short- timescale distributions are taken into consideration. We derive efficient online inference procedures to sequentially update the model with newly arrived data and show the effectiveness of our proposed model on the Mozilla add-on reviews crawled between 2007 and 2011.

AB - Social media data are produced continuously by a large and uncontrolled number of users. The dynamic nature of such data requires the sentiment and topic analysis model to be also dynamically updated, capturing the most recent language use of sentiments and topics in text. We propose a dynamic Joint Sentiment-Topic model (dJST) which allows the detection and tracking of views of current and recurrent interests and shifts in topic and sentiment. Both topic and sentiment dynamics are captured by assuming that the current sentiment-topic-specific word distributions are generated according to the word distributions at previous epochs. We study three different ways of accounting for such dependency information: (1) sliding window where the current sentiment-topic word distributions are dependent on the previous sentiment-topic-specific word distributions in the last S epochs; (2) skip model where history sentiment topic word distributions are considered by skipping some epochs in between; and (3) multiscale model where previous long- and short- timescale distributions are taken into consideration. We derive efficient online inference procedures to sequentially update the model with newly arrived data and show the effectiveness of our proposed model on the Mozilla add-on reviews crawled between 2007 and 2011.

U2 - 10.1145/2542182.2542188

DO - 10.1145/2542182.2542188

M3 - Article

VL - 5

JO - ACM Transactions on Intelligent Systems and Technology

JF - ACM Transactions on Intelligent Systems and Technology

SN - 2157-6904

IS - 1

M1 - 6

ER -