File Download
There are no files associated with this item.
Links for fulltext
(May Require Subscription)
- Publisher Website: 10.1016/j.future.2017.04.015
- Scopus: eid_2-s2.0-85018301222
- WOS: WOS:000413060500006
- Find via
Supplementary
- Citations:
- Appears in Collections:
Article: Modeling and inferring mobile phone users’ negative emotion spreading in social networks
Title | Modeling and inferring mobile phone users’ negative emotion spreading in social networks |
---|---|
Authors | |
Keywords | Dynamic social networks Emotion contamination Latent Dirichlet Allocation Gibbs sampling |
Issue Date | 2018 |
Citation | Future Generation Computer Systems, 2018, v. 78 pt. 3, p. 933-942 How to Cite? |
Abstract | © 2017 Elsevier B.V. Individual emotion, as an important part of personal privacy on health information, is vital for physical and emotional well-being. Despite the physiological reasons, emotion contagion between peoples is pivotal to understand people's emotional changes. However, most existing works at the individual level focus on small groups in the short term. Because negative emotions are natural to appear and can largely affect the dynamics of emotion spreading, therefore this paper aims to investigate the negative emotion spreading mechanism at the individual level of large user groups in the long term, and finally infer individuals’ ability of the negative emotion spreading by observing people's behaviors on mobile social networking. Specifically, we first propose a novel metric for measuring individuals’ degree of negative emotion spreading. We then put forward a Graph-Coupled Hidden Markov Sentiment Model for modeling the propagation of infectious negative sentiment locally within a social network using data collected by mobile phones. In this model, we assume that one can infect others even if he/she is not infected, which is an extension of the traditional assumption in epidemic spreading. Because the proposed model involves parameters, to infer those parameters, Gibbs sampling method is employed. Experiments on both synthetic and real-world network datasets are carried out, and the efficacy of our proposed model is verified. The case study on real-world, as a potential application, demonstrates that the proposed model provides a useful insight for understanding the correlation between network structure and the emotion shift. |
Persistent Identifier | http://hdl.handle.net/10722/296144 |
ISSN | 2023 Impact Factor: 6.2 2023 SCImago Journal Rankings: 1.946 |
ISI Accession Number ID |
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Du, Zhanwei | - |
dc.contributor.author | Yang, Yongjian | - |
dc.contributor.author | Cai, Qing | - |
dc.contributor.author | Zhang, Chijun | - |
dc.contributor.author | Bai, Yuan | - |
dc.date.accessioned | 2021-02-11T04:52:56Z | - |
dc.date.available | 2021-02-11T04:52:56Z | - |
dc.date.issued | 2018 | - |
dc.identifier.citation | Future Generation Computer Systems, 2018, v. 78 pt. 3, p. 933-942 | - |
dc.identifier.issn | 0167-739X | - |
dc.identifier.uri | http://hdl.handle.net/10722/296144 | - |
dc.description.abstract | © 2017 Elsevier B.V. Individual emotion, as an important part of personal privacy on health information, is vital for physical and emotional well-being. Despite the physiological reasons, emotion contagion between peoples is pivotal to understand people's emotional changes. However, most existing works at the individual level focus on small groups in the short term. Because negative emotions are natural to appear and can largely affect the dynamics of emotion spreading, therefore this paper aims to investigate the negative emotion spreading mechanism at the individual level of large user groups in the long term, and finally infer individuals’ ability of the negative emotion spreading by observing people's behaviors on mobile social networking. Specifically, we first propose a novel metric for measuring individuals’ degree of negative emotion spreading. We then put forward a Graph-Coupled Hidden Markov Sentiment Model for modeling the propagation of infectious negative sentiment locally within a social network using data collected by mobile phones. In this model, we assume that one can infect others even if he/she is not infected, which is an extension of the traditional assumption in epidemic spreading. Because the proposed model involves parameters, to infer those parameters, Gibbs sampling method is employed. Experiments on both synthetic and real-world network datasets are carried out, and the efficacy of our proposed model is verified. The case study on real-world, as a potential application, demonstrates that the proposed model provides a useful insight for understanding the correlation between network structure and the emotion shift. | - |
dc.language | eng | - |
dc.relation.ispartof | Future Generation Computer Systems | - |
dc.subject | Dynamic social networks | - |
dc.subject | Emotion contamination | - |
dc.subject | Latent Dirichlet Allocation | - |
dc.subject | Gibbs sampling | - |
dc.title | Modeling and inferring mobile phone users’ negative emotion spreading in social networks | - |
dc.type | Article | - |
dc.description.nature | link_to_subscribed_fulltext | - |
dc.identifier.doi | 10.1016/j.future.2017.04.015 | - |
dc.identifier.scopus | eid_2-s2.0-85018301222 | - |
dc.identifier.volume | 78 | - |
dc.identifier.issue | pt. 3 | - |
dc.identifier.spage | 933 | - |
dc.identifier.epage | 942 | - |
dc.identifier.isi | WOS:000413060500006 | - |
dc.identifier.issnl | 0167-739X | - |