Knowledge Discovery in Social Media: Physical World, Online World, and Virtual World

Persistent Link:
http://hdl.handle.net/10150/145420
Title:
Knowledge Discovery in Social Media: Physical World, Online World, and Virtual World
Author:
Zhang, Yulei
Issue Date:
2011
Publisher:
The University of Arizona.
Rights:
Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.
Abstract:
Social media have grown tremendously, making the Internet a new platform for community-based social interaction. The rich and vast amount of social media data provides valuable resources for understanding various social phenomena. Different from the world where people physically live, the new media bring additional types of worlds into people's lives: online worlds and virtual worlds. Examples of online worlds include Web forums, blogs, and online reviews, while the most famous example of a virtual world is Second Life. My dissertation is trying to address the overarching questions about how people adapt to social media to share information and exchange opinions, and what factors influence their activities in the new media. I adopt Web mining, machine learning, and computational linguistics techniques to analyze aspects of people and their behavior, such as gender differences, emotional differences, avatar activity differences, and avatar social interaction differences in online and virtual worlds.Chapter 2 develops a feature-based text classification framework to examine online gender differences between Web forum posters by analyzing writing styles and topics of interest. Guided by the stereotyping and social roles theories, Chapter 3 examines the emotional differences between men and women in text-based online communications. A research framework for automatic emotion detection is developed using sentiment analysis techniques. In the framework, different algorithms are developed to analyze the sentence-level subjectivity and phrase- and word-level polarity. Chapters 4 and 5 focus on investigating avatar behavior in the virtual world. Guided by the theories of social presence, social role, and gender role, Chapter 4 examines the effects of avatar virtual gender, virtual age, and region theme on avatars' physical activities. Chapter 5 further examines avatars' gender and age differences in their social interactions in help-seeking regions in the virtual world. The overall gender and age difference analyses and detailed investigations by comparing three types of interaction networks based on gender or age are conducted.Overall, my dissertation contributes to the literature on social media analytics, knowledge discovery, virtual world research, and text and Web mining.
Type:
Electronic Dissertation; text
Degree Name:
Ph.D.
Degree Level:
doctoral
Degree Program:
Graduate College; Management Information Systems
Degree Grantor:
University of Arizona
Advisor:
Chen, Hsinchun

Full metadata record

DC FieldValue Language
dc.language.isoenen_US
dc.titleKnowledge Discovery in Social Media: Physical World, Online World, and Virtual Worlden_US
dc.creatorZhang, Yuleien_US
dc.contributor.authorZhang, Yuleien_US
dc.date.issued2011-
dc.publisherThe University of Arizona.en_US
dc.rightsCopyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.en_US
dc.description.abstractSocial media have grown tremendously, making the Internet a new platform for community-based social interaction. The rich and vast amount of social media data provides valuable resources for understanding various social phenomena. Different from the world where people physically live, the new media bring additional types of worlds into people's lives: online worlds and virtual worlds. Examples of online worlds include Web forums, blogs, and online reviews, while the most famous example of a virtual world is Second Life. My dissertation is trying to address the overarching questions about how people adapt to social media to share information and exchange opinions, and what factors influence their activities in the new media. I adopt Web mining, machine learning, and computational linguistics techniques to analyze aspects of people and their behavior, such as gender differences, emotional differences, avatar activity differences, and avatar social interaction differences in online and virtual worlds.Chapter 2 develops a feature-based text classification framework to examine online gender differences between Web forum posters by analyzing writing styles and topics of interest. Guided by the stereotyping and social roles theories, Chapter 3 examines the emotional differences between men and women in text-based online communications. A research framework for automatic emotion detection is developed using sentiment analysis techniques. In the framework, different algorithms are developed to analyze the sentence-level subjectivity and phrase- and word-level polarity. Chapters 4 and 5 focus on investigating avatar behavior in the virtual world. Guided by the theories of social presence, social role, and gender role, Chapter 4 examines the effects of avatar virtual gender, virtual age, and region theme on avatars' physical activities. Chapter 5 further examines avatars' gender and age differences in their social interactions in help-seeking regions in the virtual world. The overall gender and age difference analyses and detailed investigations by comparing three types of interaction networks based on gender or age are conducted.Overall, my dissertation contributes to the literature on social media analytics, knowledge discovery, virtual world research, and text and Web mining.en_US
dc.typeElectronic Dissertationen_US
dc.typetexten_US
thesis.degree.namePh.D.en_US
thesis.degree.leveldoctoralen_US
thesis.degree.disciplineGraduate Collegeen_US
thesis.degree.disciplineManagement Information Systemsen_US
thesis.degree.grantorUniversity of Arizonaen_US
dc.contributor.advisorChen, Hsinchunen_US
dc.contributor.committeememberBrown, Susan Aen_US
dc.contributor.committeememberNunamaker, Jay Fen_US
dc.identifier.proquest11548-
dc.identifier.oclc752261412-
All Items in UA Campus Repository are protected by copyright, with all rights reserved, unless otherwise indicated.