Combined Word and Network Embeddings: An Analysis Framework of User Opinions on Social Media

Search for this publication on Google Scholar
Singh, T. (2020). Combined Word and Network Embeddings: An Analysis Framework of User Opinions on Social Media. Unc Charlotte Electronic Theses And Dissertations.
Download PDF

Analytics

63 views ◎
40 downloads ⇓


Abstract

Online social media like Twitter, Facebook, and Gab is often used as the stage to deliver one's opinion for a particular group of people, a political party, etc. Sometimes, the opinions shared are considered as controversial by some audience, applauded by some, or disagreed by some in the form of comments, sharing, likes, or dislikes. The information about shared opinion and the reaction to it in the form of positive, or negative reaction forms an interaction, and a collection of many such interactions forms a signed network. In addition, the evolution of information on social networks strongly relies on the nature of interactions between the users. The study of interactions is, therefore, crucial to predict the extent and nature of information spread. In this work, we study the relationship between users whether they agree or disagree in the dynamic evolution of interactions (cascades) on a larger network, Gab, to predict the relationship between the users on the social network. We quantitatively use the combination of text information and network information to enhance state of the art deep learning models for contradiction detection. The outcome of this research might contribute to improving link prediction.

Details
Author

Singh, Tannu

Title

Combined Word and Network Embeddings: An Analysis Framework of User Opinions on Social Media

Physical Description

1 online resource (55 pages) : PDF

Date

2020

Degree Granting Institution

University of North Carolina at Charlotte

Abstract

Online social media like Twitter, Facebook, and Gab is often used as the stage to deliver one's opinion for a particular group of people, a political party, etc. Sometimes, the opinions shared are considered as controversial by some audience, applauded by some, or disagreed by some in the form of comments, sharing, likes, or dislikes. The information about shared opinion and the reaction to it in the form of positive, or negative reaction forms an interaction, and a collection of many such interactions forms a signed network. In addition, the evolution of information on social networks strongly relies on the nature of interactions between the users. The study of interactions is, therefore, crucial to predict the extent and nature of information spread. In this work, we study the relationship between users whether they agree or disagree in the dynamic evolution of interactions (cascades) on a larger network, Gab, to predict the relationship between the users on the social network. We quantitatively use the combination of text information and network information to enhance state of the art deep learning models for contradiction detection. The outcome of this research might contribute to improving link prediction.

Genre

masters theses

Subjects--Topics

Computer science

Degree

M.S.

Keywords

Combined
Dynamic
Embeddings
Networks

Subject Area

Computer Science

Advisor(s)

Krishnan, Siddharth

Committee Members

Shaikh, Samira
Terejanu, Gabriel

Degree Note

Thesis (M.S.)--University of North Carolina at Charlotte, 2020.

Rights Statement

This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s). For additional information, see http://rightsstatements.org/page/InC/1.0/.

Rights Holder Information

Copyright is held by the author unless otherwise indicated.

Identifier

Singh_uncc_0694N_12556

Permalink

http://hdl.handle.net/20.500.13093/etd:2127