Suchergebnisse

Despite their popularity and potential to promote health in large populations, the effectiveness of online social networks (e.g., Facebook) to improve health behaviors has been somewhat disappointing. Most of the research examining the effectiveness of such interventions has used randomized controlled trials (RCTs). It is asserted that the modest outcomes may be due to characteristics specific to both online social networks and RCTs. The highly controlled nature of RCTs stifles the dynamic nature of online social networks. Alternative and ecologically valid research designs that evaluate online social networks in real-life conditions are needed to advance the science in this area.

Online social networks are constantly growing in popularity. They enable users to interact with one another and shifting their relations to the virtual world. Users utilize social media platforms as a mean for a rich variety of activities. Indeed, users are able to express their opinions, share experiences, react to other users' views and exchange ideas. Such online human interactions take place within a dynamic hierarchy where we can observe and distinguish many qualities related to relations between users, concerning influential, trusted or popular individuals. In particular, influence within Social Networks (SN) has been a recent focus in the literature. Many domains, such as recommender systems or Social Network Analysis (SNA), measure and exploit users' influence. Therefore, models discovering and estimating influence are important for current research and are useful in various disciplines, such as marketing, political and social campaigns, recommendations and others. Interestingly, interactions between users can not only indicate influence but also involve trust, popularity or reputation of users. However, all these notions are still vaguely defined and not meeting the consensus in the SNA community. Defining, distinguishing and measuring the strength of those relations between the users are also posing numerous challenges, on theoretical and practical ground, and are yet to be explored. Modelization of influence poses multiple challenges. In particular, current state-of-the-art methods of influence discovery and evaluation still do not fully explore users' actions of various types, and are not adaptive enough for using different SN. Furthermore, adopting the time aspect into influence model is important, challenging and in need of further examination part of the research. Finally, exploring possible connections and links between coinciding notions, like influence and reputation, remains to be performed.In this thesis, we focus on the qualities of users connected to four important concepts: influence, ...

Online social networks are constantly growing in popularity. They enable users to interact with one another and shifting their relations to the virtual world. Users utilize social media platforms as a mean for a rich variety of activities. Indeed, users are able to express their opinions, share experiences, react to other users' views and exchange ideas. Such online human interactions take place within a dynamic hierarchy where we can observe and distinguish many qualities related to relations between users, concerning influential, trusted or popular individuals. In particular, influence within Social Networks (SN) has been a recent focus in the literature. Many domains, such as recommender systems or Social Network Analysis (SNA), measure and exploit users' influence. Therefore, models discovering and estimating influence are important for current research and are useful in various disciplines, such as marketing, political and social campaigns, recommendations and others. Interestingly, interactions between users can not only indicate influence but also involve trust, popularity or reputation of users. However, all these notions are still vaguely defined and not meeting the consensus in the SNA community. Defining, distinguishing and measuring the strength of those relations between the users are also posing numerous challenges, on theoretical and practical ground, and are yet to be explored. Modelization of influence poses multiple challenges. In particular, current state-of-the-art methods of influence discovery and evaluation still do not fully explore users' actions of various types, and are not adaptive enough for using different SN. Furthermore, adopting the time aspect into influence model is important, challenging and in need of further examination part of the research. Finally, exploring possible connections and links between coinciding notions, like influence and reputation, remains to be performed.In this thesis, we focus on the qualities of users connected to four important concepts: influence, ...

With the rapid rise of the use of Online Social Networks, people have been sharing their opinions and feelings on the Internet: they write about their personal interests and political opinions, but also about their feelings about noisy activities and sounds they hear during their daily life. This textual information could provide policy makers and city managers with insights about the community response towards specific noisy events in cities that may be useful for improving the management of these activities. In this paper, we present a methodology to analyze automatically these Internet opinions by using machine learning and Natural Language processing Technologies. This approach has allowed us to build a system that automatically detects and classifies noise complaints by source, using texts written on online social networks as input. We also present a noise-event alarm system based on statistical process control theory that uses the power of our methodology to detect problematic noise events, as well as the reason why those events caused annoyance to population.

Acknowledgments -- Introduction -- Background -- Motivation -- Challenges -- Our Solutions and Applications -- Outline of this Book -- Data Gathering and Completion -- User Accounts Alignment -- Missing Data Problems -- Matrix Factorization for Data Completion -- Multiple Social Networks Data Completion -- Summary -- Multi-source Mono-task Learning -- Application: Volunteerism Tendency Prediction -- Related Work -- Volunteerism and Personality Analysis -- Multi-view Learning with Missing Data -- Multiple Social Network Learning -- Notation -- Problem Formulations -- Optimization

Intro -- Preface -- Acknowledgements -- Contents -- About the Authors -- Abstract -- 1 Introduction to Learning Automata Models -- 1.1 Introduction -- 1.2 Learning Automata -- 1.2.1 Environment -- 1.2.2 Automaton -- 1.2.3 Fixed-Structure Learning Automata (FSLA) -- 1.2.4 Variable-Structure Learning Automata (VSLA) -- 1.2.5 Variable Action Set Learning Automata -- 1.2.6 Continuous Action-Set Learning Automata (CALA) -- 1.2.7 Non-estimator Learning Algorithms -- 1.2.8 Estimator Learning Algorithms -- 1.2.9 Pursuit Algorithms -- 1.3 Interconnected Learning Automata -- 1.3.1 Hierarchical Structure Learning Automata (HSLA) -- 1.3.2 Multi-level Game of Learning Automata -- 1.3.3 Network of Learning Automata (NLA) -- 1.3.4 Distributed Learning Automata (DLA) -- 1.3.5 Extended Distributed Learning Automata (eDLA) -- 1.4 Recent Applications of Learning Automata -- 1.5 Conclusion -- References -- 2 Wavefront Cellular Learning Automata: A New Learning Paradigm -- 2.1 Introduction -- 2.1.1 Cellular Learning Automata -- 2.2 Wavefront Cellular Learning Automata -- 2.2.1 Analysis -- 2.3 Conclusion -- References -- 3 Social Networks and Learning Systems: A Bibliometric Analysis -- 3.1 Introduction -- 3.2 Material and Method -- 3.2.1 Data Collection and Initial Results -- 3.2.2 Refining the Initial Results -- 3.2.3 Analyzing the Final Results -- 3.3 Results -- 3.3.1 Initial Result Statistics -- 3.3.2 Key Journals -- 3.3.3 Key Researchers -- 3.3.4 Key Articles -- 3.3.5 Key Affiliation -- 3.3.6 Top Keywords -- 3.4 Conclusion -- References -- 4 Social Network Sampling -- 4.1 Introduction -- 4.2 Categorization of Graph Sampling Algorithms -- 4.2.1 Random Versus Topology-Based Sampling -- 4.2.2 Simple Versus Extended Sampling -- 4.2.3 Static Versus Streaming Graph Sampling -- 4.3 Learning Automata Based Graph Sampling Algorithms.

The present volume provides?a comprehensive resource for practitioners and researchers alike-both those new to the field as well as those who already have some experience. The work covers Social Network Analysis theory and methods with a focus on current applications and case studies applied in various domains such as mobile networks, security, machine learning and health.?With the increasing popularity of Web 2.0, social media has become a widely used communication platform. Parallel to this development,?Social Network Analysis gained in importance as a research field, while opening up many opportunities in different application domains. Forming a bridge between theory and applications makes this work appealing to both academics and practitioners as well as graduate students.

Intro -- Contents -- Social Network to Improve the Educational Experience with the Deployment of Different Learning Models -- 1 Introduction -- 2 Social Networks in Education -- 2.1 Facebook -- 2.2 Twitter -- 2.3 YouTube -- 3 SLNs: Sporadic Learning Networks -- 4 OPPIA Platform -- 4.1 Layer Model -- 4.2 OPPIA Architecture -- 4.3 OPPIA Operation -- 5 OPPIA Implementation -- 6 Conclusions and Future Work -- References -- Temporal Model of the Online Customer Review Helpfulness Prediction with Regression Methods -- 1 Introduction -- 2 Related Works -- 2.1 Linear Regression -- 2.2 The Coefficient of Determination -- 2.3 The Akaike Information Criterion -- 3 Method -- 3.1 Corpus Collection -- 3.2 Morphological Preprocessing -- 3.3 Feature Set -- 3.4 Sentiment Feature Selection -- 3.5 Evaluation Index -- 4 Experiments -- 4.1 Authors and Affiliations of Chinese Customer Review Corpus -- 4.2 Experimental Tools -- 4.3 Experimental Results -- 4.4 Discussion -- 5 Conclusion and Future Works -- References -- Traits of Leaders in Movement Initiation: Classification and Identification -- 1 Introduction -- 2 The Proposed Approach -- 2.1 Bidirectional Agreement in Multi-Agent Systems -- 2.2 Bidirectional Agreement Condition -- 2.3 Leaders as State Changers -- 2.4 Approach Overview -- 2.5 FLICA -- 2.6 Leadership Trait Characterization Scheme -- 3 Experimental Setup -- 3.1 Trait of Leadership Model -- 3.2 Datasets -- 3.3 Sensitivity Analysis in Model Classification -- 3.4 Hypotheses Tests -- 3.5 Parameter Setting -- 4 Results -- 4.1 Traits of Leader Classification: Sensitivity Analysis -- 4.2 Trait Identification of Baboon Movement -- 4.3 Trait Identification of Fish Movement -- 4.4 Traits of Leaders as Measure of Degree of Hierarchy Structure -- 5 Conclusions -- References -- Emotional Valence Shifts and User Behavior on Twitter, Facebook, and YouTube.

This study, using in-depth interviews and focus groups, examines perceptions of social networking sites as a means of communicating with Generation Z, from the perspectives of the major Irish political parties using these online resources and the perspective of their young target audience. There are two research questions: (1) How do political parties perceive social networking sites' role in communicating with Generation Z? and (2) How do members of Generation Z perceive social networking sites' role in communicating with political parties?

Social media platforms such as social network sites (SNS) have become an essential part of users' everyday activities. Users frequently engage in SNS to socialise and access information. The business model of several SNS firms is rooted in advertising. The challenge for SNS firms is to make sure that users accept SNS advertising. Based on uses and gratifications theory (UGT) and motivated reasoning theory (MRT), the current research offers a motivational cognitive mechanism model of SNS advertising acceptance. To assess the applicability of this model, the current research assesses how pre-purchase and ongoing information seeking motivations via cognitive processes influence users' SNS banner ad click behaviour. To attest the offered theory and hypotheses, an offline survey using self-administered questionnaire was conducted with 240 SNS users living in Pakistan. Data supported the proposed model and indicated that pre-purchase and ongoing SNS information seeking motivations via cognitive processes influence users' SNS banner ad click behaviour.