Topic evolution modeling in social media short texts based on recurrent semantic dependent CRP

Yuhao Zhang, Wenji Mao, Dajun Zeng

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

1 Citation (Scopus)

Abstract

Social media has become an important platform for people to express opinions, share information and communicate with others. Detecting and tracking topics from social media can help people grasp essential information and facilitate many security-related applications. As social media texts are usually short, traditional topic evolution models built based on LDA or HDP often suffer from the data sparsity problem. Recently proposed topic evolution models are more suitable for short texts, but they need to manually specify topic number which is fixed during different time period. To address these issues, in this paper, we propose a nonparametric topic evolution model for social media short texts. We first propose the recurrent semantic dependent Chinese restaurant process (rsdCRP), which is a nonparametric process incorporating word embeddings to capture semantic similarity information. Then we combine rsdCRP with word co-occurrence modeling and build our short-Text oriented topic evolution model sdTEM. We carry out experimental studies on Twitter dataset. The results demonstrate the effectiveness of our method to monitor social media topic evolution compared to the baseline methods.

Original languageEnglish (US)
Title of host publication2017 IEEE International Conference on Intelligence and Security Informatics
Subtitle of host publicationSecurity and Big Data, ISI 2017
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages119-124
Number of pages6
ISBN (Electronic)9781509067275
DOIs
StatePublished - Aug 8 2017
Externally publishedYes
Event15th IEEE International Conference on Intelligence and Security Informatics, ISI 2017 - Beijing, China
Duration: Jul 22 2017Jul 24 2017

Other

Other15th IEEE International Conference on Intelligence and Security Informatics, ISI 2017
CountryChina
CityBeijing
Period7/22/177/24/17

Fingerprint

Semantics
Social media
Modeling
Chinese restaurant

Keywords

  • Social Media Analytics
  • Text Mining
  • Topic Modeling

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Networks and Communications
  • Information Systems
  • Information Systems and Management
  • Safety, Risk, Reliability and Quality

Cite this

Zhang, Y., Mao, W., & Zeng, D. (2017). Topic evolution modeling in social media short texts based on recurrent semantic dependent CRP. In 2017 IEEE International Conference on Intelligence and Security Informatics: Security and Big Data, ISI 2017 (pp. 119-124). [8004885] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ISI.2017.8004885

Topic evolution modeling in social media short texts based on recurrent semantic dependent CRP. / Zhang, Yuhao; Mao, Wenji; Zeng, Dajun.

2017 IEEE International Conference on Intelligence and Security Informatics: Security and Big Data, ISI 2017. Institute of Electrical and Electronics Engineers Inc., 2017. p. 119-124 8004885.

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

Zhang, Y, Mao, W & Zeng, D 2017, Topic evolution modeling in social media short texts based on recurrent semantic dependent CRP. in 2017 IEEE International Conference on Intelligence and Security Informatics: Security and Big Data, ISI 2017., 8004885, Institute of Electrical and Electronics Engineers Inc., pp. 119-124, 15th IEEE International Conference on Intelligence and Security Informatics, ISI 2017, Beijing, China, 7/22/17. https://doi.org/10.1109/ISI.2017.8004885
Zhang Y, Mao W, Zeng D. Topic evolution modeling in social media short texts based on recurrent semantic dependent CRP. In 2017 IEEE International Conference on Intelligence and Security Informatics: Security and Big Data, ISI 2017. Institute of Electrical and Electronics Engineers Inc. 2017. p. 119-124. 8004885 https://doi.org/10.1109/ISI.2017.8004885
Zhang, Yuhao ; Mao, Wenji ; Zeng, Dajun. / Topic evolution modeling in social media short texts based on recurrent semantic dependent CRP. 2017 IEEE International Conference on Intelligence and Security Informatics: Security and Big Data, ISI 2017. Institute of Electrical and Electronics Engineers Inc., 2017. pp. 119-124
@inproceedings{aca756fd4e054e95a04a54bc4dd202a4,
title = "Topic evolution modeling in social media short texts based on recurrent semantic dependent CRP",
abstract = "Social media has become an important platform for people to express opinions, share information and communicate with others. Detecting and tracking topics from social media can help people grasp essential information and facilitate many security-related applications. As social media texts are usually short, traditional topic evolution models built based on LDA or HDP often suffer from the data sparsity problem. Recently proposed topic evolution models are more suitable for short texts, but they need to manually specify topic number which is fixed during different time period. To address these issues, in this paper, we propose a nonparametric topic evolution model for social media short texts. We first propose the recurrent semantic dependent Chinese restaurant process (rsdCRP), which is a nonparametric process incorporating word embeddings to capture semantic similarity information. Then we combine rsdCRP with word co-occurrence modeling and build our short-Text oriented topic evolution model sdTEM. We carry out experimental studies on Twitter dataset. The results demonstrate the effectiveness of our method to monitor social media topic evolution compared to the baseline methods.",
keywords = "Social Media Analytics, Text Mining, Topic Modeling",
author = "Yuhao Zhang and Wenji Mao and Dajun Zeng",
year = "2017",
month = "8",
day = "8",
doi = "10.1109/ISI.2017.8004885",
language = "English (US)",
pages = "119--124",
booktitle = "2017 IEEE International Conference on Intelligence and Security Informatics",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
address = "United States",

}

TY - GEN

T1 - Topic evolution modeling in social media short texts based on recurrent semantic dependent CRP

AU - Zhang, Yuhao

AU - Mao, Wenji

AU - Zeng, Dajun

PY - 2017/8/8

Y1 - 2017/8/8

N2 - Social media has become an important platform for people to express opinions, share information and communicate with others. Detecting and tracking topics from social media can help people grasp essential information and facilitate many security-related applications. As social media texts are usually short, traditional topic evolution models built based on LDA or HDP often suffer from the data sparsity problem. Recently proposed topic evolution models are more suitable for short texts, but they need to manually specify topic number which is fixed during different time period. To address these issues, in this paper, we propose a nonparametric topic evolution model for social media short texts. We first propose the recurrent semantic dependent Chinese restaurant process (rsdCRP), which is a nonparametric process incorporating word embeddings to capture semantic similarity information. Then we combine rsdCRP with word co-occurrence modeling and build our short-Text oriented topic evolution model sdTEM. We carry out experimental studies on Twitter dataset. The results demonstrate the effectiveness of our method to monitor social media topic evolution compared to the baseline methods.

AB - Social media has become an important platform for people to express opinions, share information and communicate with others. Detecting and tracking topics from social media can help people grasp essential information and facilitate many security-related applications. As social media texts are usually short, traditional topic evolution models built based on LDA or HDP often suffer from the data sparsity problem. Recently proposed topic evolution models are more suitable for short texts, but they need to manually specify topic number which is fixed during different time period. To address these issues, in this paper, we propose a nonparametric topic evolution model for social media short texts. We first propose the recurrent semantic dependent Chinese restaurant process (rsdCRP), which is a nonparametric process incorporating word embeddings to capture semantic similarity information. Then we combine rsdCRP with word co-occurrence modeling and build our short-Text oriented topic evolution model sdTEM. We carry out experimental studies on Twitter dataset. The results demonstrate the effectiveness of our method to monitor social media topic evolution compared to the baseline methods.

KW - Social Media Analytics

KW - Text Mining

KW - Topic Modeling

UR - http://www.scopus.com/inward/record.url?scp=85030233335&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85030233335&partnerID=8YFLogxK

U2 - 10.1109/ISI.2017.8004885

DO - 10.1109/ISI.2017.8004885

M3 - Conference contribution

AN - SCOPUS:85030233335

SP - 119

EP - 124

BT - 2017 IEEE International Conference on Intelligence and Security Informatics

PB - Institute of Electrical and Electronics Engineers Inc.

ER -